en - ec.europa.euec.europa.eu/.../consultations/2010_horizontals/guidelines_en.pdf · en 2 en draft...

EN EN

EN

EN EN

EUROPEAN COMMISSION

Brussels, SEC(2010) 528/2

Draft

COMMUNICATION FROM THE COMMISSION

Guidelines on the applicability of Article 101 of the Treaty on the Functioning of the European Union to horizontal co-operation agreements

EN 2 EN

Draft

COMMUNICATION FROM THE COMMISSION

Guidelines on the applicability of Article 101 of the Treaty on the Functioning of the European Union to horizontal co-operation agreements

(Text with EEA relevance)

TABLE OF CONTENTS

1. Introduction.................................................................................................................. 5

1.1. Purpose and scope ........................................................................................................ 5

1.2. Basic principles for the assessment under Article 101................................................. 9

1.2.1 Article 101(1) ............................................................................................................. 10

1.2.2. Article 101(3) ............................................................................................................. 14

1.3. Structure of these guidelines ...................................................................................... 15

2. General Principles on the competitive assessment of information exchange ............ 16

2.1. Definition and scope .................................................................................................. 16

2.2. Assessment under Article 101(1) ............................................................................... 17

2.2.1. Main competition concerns ........................................................................................ 17

2.2.2. Restriction of competition by object .......................................................................... 18

2.2.3. Restrictive effects on competition.............................................................................. 19


2.3.1. Efficiency gains.......................................................................................................... 24

2.3.2. Indispensability .......................................................................................................... 25

2.3.3. Pass-on to consumers ................................................................................................. 25

2.3.4. No elimination of competition ................................................................................... 26

2.4. Examples .................................................................................................................... 26

3. Research and Development Agreements ................................................................... 30

3.1. Definition ................................................................................................................... 30

3.2. Relevant markets........................................................................................................ 30


EN 3 EN


3.3.2. Restrictions of competition by object ........................................................................ 33







3.4.5. Time of the assessment .............................................................................................. 37

3.5. Examples .................................................................................................................... 38

4. Production Agreements.............................................................................................. 40

4.1. Definition and scope .................................................................................................. 40











4.5. Examples .................................................................................................................... 47

5. Purchasing agreements............................................................................................... 52

5.1. Definition ................................................................................................................... 52







EN 4 EN





5.5. Examples .................................................................................................................... 56

6. Agreements on Commercialisation ............................................................................ 58

6.1. Definition ................................................................................................................... 58











6.5. Examples .................................................................................................................... 63

7. Standardisation Agreements....................................................................................... 66

7.1. Definition ................................................................................................................... 66











7.5. Examples .................................................................................................................... 76

EN 5 EN

1. INTRODUCTION

1.1. Purpose and scope

1. These guidelines set out the principles for the assessment under Article 101 of the Treaty on the Functioning of the European Union ("Article 101") of agreements between undertakings, decisions of associations of undertakings and concerted practices (collectively referred to as "agreements") pertaining to horizontal co-operations. A co-operation is of a 'horizontal nature' if an agreement is entered into between actual or potential competitors. In addition, these guidelines also cover horizontal co-operation agreements between non-competitors, e.g., between two companies active in the same product markets but active in different geographic markets without being potential competitors.

2. Horizontal co-operation agreements can lead to substantial economic benefits, in particular if they combine complementary activities, skills or assets. Horizontal co-operation can be a means to share risk, save costs, increase investments, pool know-how, enhance product quality and variety, and launch innovation faster.

3. On the other hand, horizontal co-operation agreements may lead to competition problems. This is for example the case if the parties agree to fix prices or output, to share markets, or if the co-operation enables the parties to maintain, gain or increase market power and thereby is likely to give rise to negative market effects with respect to prices, output, innovation or the variety and quality of products.

4. The Commission, while recognising the benefits that can be generated by horizontal co-operation agreements, has to ensure that effective competition is maintained. Article 101 provides the legal framework for a balanced assessment taking into account both restrictive effects on competition as well as pro-competitive effects.

5. The purpose of these guidelines is to provide an analytical framework for the most common types of horizontal co-operation agreements; they deal with research and development agreements, production agreements including subcontracting and specialisation agreements, purchasing agreements, commercialisation agreements, standardisation agreements including standard contracts, and information exchange. This framework is primarily based on legal and economic criteria that help to analyse a horizontal co-operation agreement and the context in which it occurs. Economic criteria such as the market power of the parties and other factors relating to the market structure form a key element of the assessment of the market impact likely to be caused by a horizontal co-operation agreement and therefore for the assessment under Article 101.

6. These guidelines apply to the most common types of horizontal co-operation agreements irrespective of the level of integration they entail with the exception of operations constituting a concentration within the meaning of Article 3 of the Merger Regulation1 as would be the case, e.g., for full-function joint ventures2).

1 Council Regulation (EC) No 139/2004 on the control of concentrations between undertakings, OJ L24,

29.1.2004, p. 1.

EN 6 EN

7. Given the potentially large number of types and combinations of horizontal co-operation and market circumstances in which they operate, it is difficult to provide specific answers for every possible scenario. The present guidelines based on legal and economic criteria are intended to assist businesses in assessing the compatibility of an individual co-operation agreement with Article 101. These criteria do not, however, constitute a "checklist" which can be applied mechanically. Each case must be assessed on the basis of its own facts which may require a flexible application of these guidelines.

8. The criteria set out in these guidelines apply to horizontal co-operation agreements concerning both goods and services (collectively referred to as “products”). These guidelines complement the R&D Block Exemption Regulation3 and the Specialisation Block Exemption Regulation.4

9. Whereas these guidelines contain certain references to cartels, they are not intended to give any guidance as to what does and does not constitute a cartel as defined by the decisional practice of the Commission and the jurisprudence of the Court of Justice of the European Union.

10. The term "competitors" as used in these guidelines includes both actual and potential competitors. Two companies are treated as actual competitors if they are active on the same relevant market. A company is treated as a potential competitor of another company if, absent the agreement, in case of a small but permanent increase in relative prices it is likely that this first company, within a short period of time, would undertake the necessary additional investments or other necessary switching costs to enter the relevant market on which the other company is active. This assessment has to be based on reasonably objective grounds, the mere theoretical possibility to enter a market is not sufficient.5

11. Companies that form part of the same "undertaking" within the meaning of Article 101(1) are not considered to be competitors by these guidelines. Article 101 only applies to agreements between independent undertakings. When one company (the "parent company") exercises decisive influence over another company (the "subsidiary") they form a single economic entity and, hence, are part of the same undertaking.6 The same is true for sister companies, i.e., subsidiaries over which

2 See Article 3(4) of the Merger Regulation; regarding the notion of full-function joint ventures, i.e., joint

ventures that perform on a lasting basis all the functions of an autonomous economic entity, see Commission Consolidated Jurisdictional Notice, OJ C95, 16.4.2008. p. 1, paragraphs 91-109 ("Consolidated Jurisdictional Notice").

3 Commission Regulation (EC) No [...] on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of research and development agreements, OJ L[…] ("R&D Block Exemption Regulation").

4 Commission Regulation (EC) No […] on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of specialisation agreements, OJ L[…] ("Specialisation Block Exemption Regulation").

5 See Commission notice on the definition of the relevant market for the purposes of Community competition law, OJ C372, 9.12.1997, p. 5, paragraph 24 ("Market Definition Notice"); see also the Commission’s Thirteenth Report on Competition Policy, point 55 and Commission decision in Case IV/32.009, Elopak/Metal Box-Odin, OJ L209, 8.8.1990, p. 15.

6 See, e.g., Case C-73/95, Viho, [1996] ECR I-5457, paragraph 51. The exercise of decisive influence by the parent over the conduct of a subsidiary can be presumed in case of wholly-owned subsidiaries; see, e.g., Case 107/82, AEG, [1983] ECR-3151, paragraph 50; Case C-286/98 P, Stora, [2000] ECR-I 9925,

EN 7 EN

decisive influence is exercised by the same parent company. They are consequently not considered to be competitors even if they are both active on the same relevant product and geographic markets. As a joint venture forms part of one undertaking with each of the parent companies that jointly exercise decisive influence and effective control over it7, Article 101 does not apply to agreements between the parents and such a joint venture, provided the creation of the joint venture did not infringe EU competition law. Article 101 could, however, apply to agreements between the parents outside the scope of the joint venture and with regard to the agreement between the parents to create the joint venture.

12. Agreements that are entered into between undertakings operating at a different level of the production or distribution chain, i.e., vertical agreements, are in principle dealt with in the Block Exemption Regulation on Vertical Restraints8 and the Guidelines on Vertical Restraints.9 However, to the extent that vertical agreements, e.g., distribution agreements, are concluded between competitors, the effects of the agreement on the market and the possible competition problems can be similar to horizontal agreements. Therefore, vertical agreements between competitors have to be assessed according to the principles described in these guidelines.10 Should there be a need to also assess such agreements under the Block Exemption Regulation and the Guidelines on Vertical Restraints, this will be specifically set out in the relevant chapter of these guidelines. Absent such reference, only the following chapters of these guidelines will be applicable.

13. Horizontal co-operation agreements may combine different stages of co-operation, for example R&D and the production and/or commercialisation of its results. Such agreements are generally also covered by these guidelines. The most upstream indispensible building block of such an integrated co-operation determines which chapter of the present guidelines serves as the starting point of the assessment of the entire agreement in question. In general, the chapter so identified comprehensively deals with all stages of such integrated co-operations. However, the competition concerns and corresponding explanations set out in the chapters pertaining to the downstream activities of the integrated co-operation generally also guide the assessment of these activities. As regards the applicable "safe harbours", only those set out in the chapter pertaining to the "most upstream indispensible building block" will apply to the entire integrated co-operation.

paragraph 29; or Case C-97/08 P, Akzo, [2009] ECR I not yet reported, paragraphs 60 et seqq. Where there are several parent companies, the exercise of decisive influence by them can be presumed where they control all the shares in the joint venture, and where, on the basis of factual evidence, they have joint management power; see Case T-314/01, Avebe, ECR II-3085, paragraphs 138 and 139.

7 See Case T-314/01, Avebe, cited in note 6 above, paragraphs 138 and 139. Regarding the notion of control, see Commission Consolidated Jurisdictional Notice, cited in note 2 above, paragraphs 11-82.

8 Commission Regulation (EU) No 330/2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of vertical agreements and concerted practices, OJ L102, 23.4.2010, p. 2.

9 OJ C[…]. 10 This does not apply where competitors enter into a non-reciprocal vertical agreement and (i) the

supplier is a manufacturer and a distributor of goods, while the buyer is a distributor and not a competing undertaking at the manufacturing level, or (ii) the supplier is a provider of services at several levels of trade, while the buyer provides its goods or services at the retail level and is not a competing undertaking at the level of trade where it purchases the contract services. Such agreements are exclusively assessed under the Block Exemption Regulation and the Guidelines on Vertical Restraints (see Article 2(4) of the Block Exemption Regulation on Vertical Restraints, cited in note 8 above).

EN 8 EN

14. For example, the starting point of an analysis of a horizontal co-operation agreement involving both joint R&D and joint production of the results would thus normally be the chapter on “Research and Development Agreements”, as the joint production will only take place if the joint R&D is successful. This implies that the results of the joint R&D are decisive for the subsequent joint production. The R&D agreement can thus be regarded as the most upstream indispensible building block of the co-operation. This assessment would change if the parties engaged in the joint production in any event, i.e., irrespective of the joint R&D. In this case, the possible restrictive effects on competition and pro-competitive effects of the co-operation would largely relate to the joint production, and the agreement would therefore be examined according to the principles set out in the section on “Production Agreements”.

15. Article 101 only applies to those horizontal co-operation agreements which may affect trade between Member States. The following principles on the applicability of Article 101 are therefore based on the assumption that a horizontal co-operation agreement is capable of affecting trade between Member States to an appreciable extent.

16. The assessment under Article 101 as described in these guidelines is without prejudice to the possible parallel application of Article 102 of the Treaty to horizontal co-operation agreements.11

17. These guidelines are without prejudice to the interpretation the General Court and the Court of Justice of the European Union may give to the application of Article 101 to horizontal co-operation agreements.

18. These guidelines replace the Horizontal Guidelines which were published by the Commission in 200112 and do not apply to the extent that sector specific rules apply as is the case for certain agreements with regard to agriculture13, transport14 or insurance15. The Commission will continue to monitor the operation of the R&D and Specialisation Block Exemption Regulations16 and these guidelines based on market

11 See Case T-51/89, Tetra Pak I, [1990] ECR-II 309, paragraphs 25 et seqq. and Guidance on the

Commission's enforcement priorities in applying Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings, OJ C45, 24.2.2009, p. 7 ("Article 82 Guidance Paper").

12 Commission guidelines on the applicability of Article 81 of the EC Treaty to horizontal co-operation agreements, OJ C3, 6.1.2001, p. 2.

13 Council Regulation (EC) No 1184/2006 applying certain rules of competition to the production of, and trade in, agricultural products, OJ L 214, 4.8.2006, p. 7.

14 Council Regulation (EC) No 487/2009 on the application of Article 81(3) of the Treaty to certain categories of agreements and concerted practices in the air transport sector, OJ L148, 11.6.2009, p. 1; Council Regulation (EC) No 169/2009 applying rules of competition to transport by rail, road and inland waterway, OJ L61, 5.3.2009, p. 1; Council Regulation (EC) No 246/2009 on the application of Article 81(3) of the Treaty to certain categories of agreements and concerted practices between liner shipping companies (consortia), OJ L79, 25.3.2009, p. 1; Commission Regulation (EC) No 823/2000 on the application of Article 81(3) of the Treaty to certain categories of agreements, decisions and concerted practices between liner shipping companies (consortia), OJ L100, 20.4.2000, p. 24 as last amended by Commission Regulation (EC) No 611/2005, OJ L101, 20.04.2005, p. 10; Guidelines on the Article 81 of the EC Treaty to maritime transport services, OJ C245, 26.9.2008, p. 2.

15 Commission Regulation (EU) No 267/2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to certain categories of agreements, decisions and concerted practices in the insurance sector, OJ L83, 31.3.2010, p. 1.

16 Cited in notes 3 and 4 above.

EN 9 EN

information from stakeholders and national competition authorities and may revise these guidelines in light of future developments and of evolving insight.

19. The General Guidelines17 contain general guidance on the interpretation of Article 101. Consequently, the present guidelines have to be read consistently with the General Guidelines.

1.2. Basic principles for the assessment under Article 101

20. The assessment under Article 101 consists of two steps. The first step, under Article 101(1), is to assess whether an agreement between undertakings, which is capable of affecting trade between Member States, has an anti-competitive object or actual or potential18 restrictive effects on competition. The second step, under Article 101(3), which only becomes relevant when an agreement is found to be restrictive of competition within the meaning of Article 101(1), is to determine the pro-competitive benefits produced by that agreement and to assess whether these pro-competitive effects outweigh the restrictive effects on competition.19 The balancing of restrictive and pro-competitive effects is conducted exclusively within the framework laid down by Article 101(3).20 In case the pro-competitive effects do not outweigh a restriction of competition, Article 101(2) stipulates that the agreement shall be automatically void.

21. In certain cases, companies are encouraged by government authorities to enter into horizontal co-operation agreements in order to attain a public policy objective by way of self-regulation. Competition rules continue to apply if national legislation does not rule out all scope for competitive conduct by the companies concerned. In other words, the fact that government authorities encourage, endorse or sponsor a horizontal co-operation agreement does not mean that it is permissible under Article 101.21 However, agreements do not restrict competition if legislation creates a legal framework which has precluded all scope for any competitive conduct by the parties.22 Moreover, the undertakings concerned are shielded from all the consequences of an infringement of Article 101 if national legislation requires them to engage in anti-competitive conduct23 or government authorities impose irresistible

17 Commission guidelines on the application of Article 81(3) of the Treaty, OJ C101, 27.4.2004, p. 97. 18 Article 101(1) prohibits both actual and potential anti-competitive effects; see e.g. Case C-7/95 P, John

Deere, [1998] ECR I-3111, paragraph 77; Case C-238/05, Asnef-Equifax, [2006] ECR I-11125, paragraph 50.

19 See Joined Cases C-501/06 P et al, GlaxoSmithKline, [2009] ECR I not yet reported, paragraph 95. 20 See Case T-65/98, Van den Bergh Foods, [2003] ECR II-4653, paragraph 107; Case T-112/99,

Métropole télévision (M6) and others, [2001] ECR II-2459, paragraph 74; Case T-328/03, O2, [2006] ECR II-1231, paragraphs 69 et seq., where the Court of First Instance (now: the General Court) held that it is only in the precise framework of Article 101(3) that the pro- and anti-competitive aspects of a restriction may be weighed.

21 See Case C-198/01, CIF, [2003] ECR I-8055, paragraphs 56-58; Joined Cases T-217/03 and T-245/03, French Beef, [2006] ECR II-4987, paragraph 92; Case T-7/92, Asia Motor France II, [1993] ECR II-669, paragraph 71; and Case T-148/89, Tréfilunion, [1995] ECR II-1063, paragraph 118.

22 This possibility has been narrowly interpreted; see, for example, Joined Cases 209/78 et al., Van Landewyck, [1980] ECR 3125, paragraphs 130-134; Joined Cases 240/82 et al., Stichting Sigarettenindustrie, [1985] ECR 3831, paragraphs 27-29; and Joined Cases C-359/95 P and C-379/95 P, Ladbroke Racing, [1997] ECR I-6265, paragraphs 33 et seqq.

23 At least until a decision to disapply the national legislation has been adopted and that decision has become definitive; see Case C-198/01, CIF, cited in note 21 above, paragraphs 54 et seq.

EN 10 EN

pressure on them to do so.24 Each case must be assessed on its own facts according to the general principles set out below.

1.2.1 Article 101(1)

22. Article 101(1) prohibits agreements whose object or effect is to restrict25 competition.

(i) Restrictions of competition by object

23. Restrictions of competition by object are those that by their very nature have the potential of restricting competition. Further guidance with regard to the notion of restrictions of competition by object can be obtained in the General Guidelines26.

(ii) Restrictive effects on competition

24. If a horizontal co-operation agreement is not restrictive of competition by object, it must be examined whether it has appreciable restrictive effects on competition. Account must be taken of both actual and potential effects. In other words, the agreement must have at least likely anti-competitive effects.

25. For an agreement to have restrictive effects on competition within the meaning of Article 101(1) it must have an actual or likely appreciable adverse impact on at least one of the parameters of competition on the market, such as price, output, product quality, product variety or innovation. Agreements can have such effects by appreciably reducing competition between the parties to the agreement or between any one of them and third parties. This means that the agreement must reduce the parties' decision-making independence,27 either due to obligations contained in the agreement which regulate the market conduct of at least one of the parties or by influencing the market conduct of at least one of the parties by causing a change in its incentives.

26. Restrictive effects on competition within the relevant market are likely to occur where due to the agreement the parties would be able to profitably raise price or reduce output, innovation, product quality or variety. This will depend on several factors such as the nature and content of the agreement, the extent to which the parties individually or jointly have or obtain some degree of market power and the agreement contributes to the creation, maintenance or strengthening of that market power or allows the parties to exploit such market power. For restrictive effects on competition to be likely, they should be expected with a reasonable degree of probability.

27. The assessment of whether a horizontal co-operation agreement has restrictive effects on competition within the meaning of Article 101(1) must be made in

24 Case C-198/01, CIF, cited in note 21 above, paragraph 54; Case T-387/94, Asia Motor France III,

[1996] ECR II-961, paragraph 65. 25 For the purpose of these guidelines, the term “restriction of competition” includes the prevention and

distortion of competition. 26 Cited in note 17 above. 27 See Case C-7/95 P, John Deere, cited in note 18 above, paragraph 88; Case C-238/05, Asnef-Equifax,

cited in note 18 above, paragraph 51.

EN 11 EN

comparison to the actual legal and economic context in which competition would occur in the absence of the agreement with all of its alleged restrictions, (i.e., in the absence of the agreement as it stands (if already implemented) or as envisaged (if not yet implemented) at the time of assessment). Hence in order to prove actual or potential restrictive effects on competition, it is necessary to take into account competition between the parties and competition from third parties, in particular actual or potential competition that would have existed in the absence of the agreement. This comparison does not take into account any potential efficiency gains generated by the agreement as these will only be assessed under Article 101(3).

28. Consequently, horizontal co-operation agreements between competitors that would not be able to independently carry out the project or activity covered by the co-operation will normally not give rise to restrictive effects on competition within the meaning of Article 101(1) unless they could have carried out the project with less stringent restrictions.

29. General guidance with regard to the notion of restrictions of competition by effect can be obtained in the General Guidelines.28 The following paragraphs provide additional guidance specific to the assessment of horizontal co-operation agreements.

Horizontal co-operation agreements and horizontal mergers

30. The analysis of horizontal co-operation agreements has certain common elements with the analysis of horizontal mergers pertaining to the potential restrictive effects, in particular as regards joint ventures. There is often only a fine line between full function joint ventures that fall under the Merger Regulation29 and non-full function joint ventures that are assessed under Article 101. Their effects can be, however, quite similar.

Nature and content of the agreement

31. The nature and content of an agreement relates to factors such as the area and objective of the co-operation, the competitive relationship between the parties and the extent to which they combine their activities. These factors determine which kinds of possible competition concerns can arise from a horizontal co-operation agreement..

32. Horizontal co-operation agreements may limit competition in several ways. The agreement may:

– be exclusive in the sense that it limits, for instance as a result of contractual obligations, the possibility of the parties to compete against each other or third parties as independent economic operators or as parties to other, competing agreements;

– require the parties to contribute such assets that their decision-making independence is appreciably reduced; or

28 Cited in note 17 above. 29 Cited in note 1 above.

EN 12 EN

– affect the parties' financial interests in such a way that their decision-making independence is appreciably reduced. Both financial interests in the agreement but also financial interests in other parties to the agreement are relevant for the assessment.

33. The potential effect of such agreements may be the loss of competition between the parties to the agreement. Competitors can also benefit from the reduction of competitive pressure that results from the agreement and may therefore find it profitable to increase their prices. The reduction in these competitive constraints may lead to price increases in the relevant market. Factors such as whether the parties to the agreement have high market shares, whether they are close competitors, whether the customers have limited possibilities of switching suppliers, whether competitors are unlikely to increase supply if prices increase, and whether one of the parties to the agreement is an important competitive force, are all relevant for the competitive assessment of the agreement.

34. A horizontal co-operation agreement may also:

– lead to the disclosure of commercially sensitive information thereby increasing the likelihood of coordination among the parties within or outside the field of the co-operation; and/or

– cause a significant degree of commonality in costs, so the parties may more easily coordinate market prices and output. This could happen if the area of co-operation, e.g. production and purchasing, accounts for a high proportion of the parties' variable costs in a given market and the parties combine their activities in the area of co-operation to a significant extent. This could, for instance, be the case, where they jointly manufacture or purchase an important intermediate product or a high proportion of their total output of a final product.

35. A horizontal agreement may therefore decrease their decision-making independence and as a result increase the likelihood that undertakings will coordinate their behaviour in order to reach a collusive outcome but it may also make coordination easier, more stable or more effective for undertakings that were already coordinating before, either by making the coordination more robust or by permitting them to reach even higher prices.

36. Some horizontal co-operation agreements, for example production and standardisation agreements, may also give rise to anti-competitive foreclosure concerns.

Market power and other market characteristics

37. Market power is the ability to profitably maintain prices above competitive levels for a period of time or to profitably maintain output in terms of product quantities, product quality and variety or innovation below competitive levels for a period of time.

38. In markets with fixed costs undertakings must price above their variable costs of production in order to ensure a competitive return on their investment. The fact that undertakings price above their variable costs is therefore not in itself a sign that competition in the market is not functioning well and that undertakings have market

EN 13 EN

power that allows them to price above the competitive level. It is when competitive constraints are insufficient to maintain prices, output, innovation, and product quality and variety at competitive levels that undertakings have market power within the meaning of Article 101(1).

39. The creation, maintenance or strengthening of market power can result from reduced competition between the parties to an agreement. It can also result from reduced competition between any one of the parties and third parties, e.g., because the agreement leads to anti-competitive foreclosure of competitors or because it raises competitors’ costs, limiting their capacity to compete effectively with the contracting parties.

40. Market power is a question of degree. The degree of market power required for the finding of an infringement under Article 101(1) in the case of agreements that are restrictive of competition by effect is less than the degree of market power required for a finding of dominance under Article 102, where a substantial degree of market power is required.

41. The starting point for the analysis of market power is the position of the parties in the markets affected by the co-operation. To carry out this analysis the relevant market(s) have to be defined by using the methodology of the Commission's Market Definition Notice30. Where specific types of markets are concerned such as purchasing or technology markets, these guidelines will provide additional guidance.

42. If the parties have a low combined market share31, the horizontal co-operation agreement is unlikely to give rise to restrictive effects on competition within the meaning of Article 101(1) and, normally, no further analysis will be required. If one of just two parties has only an insignificant market share and if it does not possess important resources, even a high combined market share normally cannot be seen as indicating a likely restrictive effect on competition in the market.32 Given the variety of horizontal co-operation agreements and the different effects they may cause in different market situations, it is hardly possible to give a general market share threshold above which sufficient market power for causing restrictive effects on competition can be assumed.

43. Depending on the market position of the parties and the concentration in the market, other factors such as the stability of market shares over time, entry barriers and the likelihood of market entry, the countervailing power of buyers/suppliers have to be considered as well.

30 See Market definition notice, cited in note 5 above. 31 What is considered to be a "low combined market share" depends on the type of agreement in question

and can be inferred from the "safe harbour" thresholds set out in various chapters of these guidelines and, more generally, from the Commission notice on agreements of minor importance which do not appreciably restrict competition under Article 81(1) of the Treaty establish the European Community (de minimis), OJ C368, 22.12.2001, p. 13 ("De Minimis Notice").

32 If there are more than two parties, then the collective share of all co-operating competitors has to be significantly greater than the share of the largest single participating competitor.

EN 14 EN

44. Normally, the Commission uses current market shares in its competitive analysis.33 However, reasonably certain future developments may also be taken into account, for instance in the light of exit, entry or expansion in the relevant market. Historic data may be used if market shares have been volatile, for instance when the market is characterised by large, lumpy orders. Changes in historic market shares may provide useful information about the competitive process and the likely future importance of the various competitors, for instance, by indicating whether undertakings have been gaining or losing market shares. In any event, the Commission interprets market shares in the light of likely market conditions, for instance, if the market is highly dynamic in character and if the market structure is unstable due to innovation or growth.

45. When entering a market is sufficiently easy, an agreement will normally not be expected to give rise to restrictive effects on competition. For entry to be considered a sufficient competitive constraint on the parties to an agreement, it must be shown to be likely, timely and sufficient to deter or defeat any potential restrictive effects of the agreement. The analysis of entry may be affected by the presence of agreements. The likely or possible termination of an agreement may influence the likelihood of entry.

1.2.2. Article 101(3)

46. The assessment of restrictions by object or effect under Article 101(1) is only one side of the analysis. The other side, which is reflected in Article 101(3), is the assessment of the pro-competitive effects of restrictive agreements. Where in an individual case a restriction of competition within the meaning of Article 101(1) has been proven, Article 101(3) can be invoked as a defence. According to Article 2 of Regulation 1/200334 the burden of proof under Article 101(3) rests on the undertaking(s) invoking the benefit of the exception rule. Therefore, the factual arguments and the evidence provided by the undertaking(s) must enable the Commission to arrive at the conviction that the agreement in question is sufficiently likely to give rise to pro-competitive effects or that it is not.35

47. The application of the exception rule of Article 101(3) is subject to four cumulative conditions, two positive and two negative:

– the agreement must contribute to improving the production or distribution of goods or contribute to promoting technical or economic progress, i.e., lead to efficiency gains;

– the restrictions must be indispensable to the attainment of these objectives, i.e., the efficiency gains;

– consumers must receive a fair share of the resulting benefits, i.e., the efficiency gains attained by the indispensable restrictions must be sufficiently passed on to

33 As to the calculation of market shares, see also Market Definition Notice, cited in note 5 above,

paragraphs 54-55. 34 Council Regulation (EC) No 1/2003 on the implementation of the rules on competition laid down in

Articles 81 and 82 of the Treaty, OJ L1, 4.1.2003, p. 1 ("Regulation 1/2003"). 35 See, e.g., Joined Cases C-501/06 P et al., GlaxoSmithKline, cited in note 19 above, paragraphs 93-95.

EN 15 EN

consumers. Hence, efficiencies only accruing to the parties to the agreement will not suffice. For the purposes of these guidelines, the concept of “consumers” encompasses the customers, potential and/or actual, of the parties to the agreement; and

– the agreement must not afford the parties the possibility of eliminating competition in respect of a substantial part of the products in question.

48. When these four conditions are fulfilled the restrictive effects on competition generated by the agreement can be considered to be outweighed by its pro-competitive effects, thereby compensating consumers for the adverse effects of the restrictions of competition.

49. In the area of horizontal co-operation agreements there are block exemption regulations based on Article 101(3) for research and development36 and specialisation (including joint production)37 agreements. These Block Exemption Regulations are based on the premise that the combination of complementary skills or assets can be the source of substantial efficiencies in research and development and specialisation agreements. This may also be the case for other types of horizontal co-operation agreements. The analysis of the efficiencies of an individual agreement under Article 101(3) is therefore to a large extent a question of identifying the complementary skills and assets that each of the parties bring to the agreement and evaluating whether the resulting efficiencies are such that the conditions of Article 101(3) are fulfilled.

50. Complementarities may arise from horizontal co-operation agreements in various ways. A research and development agreement may bring together different research capabilities that allow the parties to produce better products cheaper and shorten the time for these products to reach the market. A production agreement may allow the parties to achieve economies of scale or scope that they could not reach individually.

51. Horizontal co-operation agreements that do not involve the combination of complementary skills or assets are less likely to lead to efficiency gains that benefit consumers. Such agreements may reduce duplication of certain costs, for instance because certain fixed costs can be eliminated. However, fixed cost savings are, in general, less likely to result in benefits to consumers than savings in, for instance, variable or marginal costs.

52. Further guidance regarding the Commission's application of the criteria of Article 101(3) can be obtained in the General Guidelines.38

1.3. Structure of these guidelines

53. The following chapters of these guidelines will first set out some general principles for the assessment of the exchange of information, which are equally applicable to all types of horizontal co-operation agreements entailing the exchange of information. The subsequent chapters of these guidelines will each address one specific type of

36 R&D Block Exemption Regulation, cited in note 3 above. 37 Specialisation Block Exemption Regulation, cited in note 4 above. 38 Cited in note 17 above.

EN 16 EN

horizontal co-operation agreement. Each chapter will apply the analytical framework described under point 1.2 above as well as the general principles on the exchange of information to the specific type of co-operation in question.

2. GENERAL PRINCIPLES ON THE COMPETITIVE ASSESSMENT OF INFORMATION EXCHANGE

2.1. Definition and scope

54. The purpose of this chapter is to guide the competitive assessment of information exchange. Information exchange can take various forms such as direct sharing of data between competitors, through a common agency (e.g., trade association), a third party39, or by means of publishing.40

55. Information exchange can only be addressed under Article 101 if it establishes or is part of an agreement, a concerted practice or a decision of an association of undertakings.

56. In line with the jurisprudence of the Court of Justice of the European Union, the concept of a concerted practice refers to a form of coordination between undertakings by which, without it having reached the stage where an agreement properly so-called has been concluded, practical cooperation between them is knowingly substituted for the risks of competition41. If there is no agreement on information exchange, it will have to be assessed on a case-by-case basis whether a concerted practice can be found or whether, for example, the regular dissemination of information of a company is a truly unilateral action which does not fall within the scope of Article 101(1)42.

57. Information exchange takes place in different contexts. There are agreements, decisions of an association of undertakings, or concerted practices under which information is exchanged where the main economic function lies in the exchange of information itself (e.g., exchanging mortality tables). Moreover, information exchange can be part of another type of horizontal agreement (e.g., the parties to a production agreement share certain information on costs). The assessment of the latter type of information exchanges should be carried out in combination with an assessment of the respective horizontal co-operation agreement.

39 Information exchange through third parties may involve, for example, indirect exchange of information

through a market research organisation or through the parties' suppliers or retailers. 40 Publishing refers to an exchange of strategic information between a group of competitors on, for

example, a common website, or to publications involving strategic responses (and readjustments of own announcements) to announcements made by competitors, which could prove to be a strategy for reaching a common understanding about the terms of coordination.

41 See for example Case C-8/08, T-Mobile Netherlands, [2009] ECR I not yet reported, paragraph 26; Joined Cases C-89/85 et.al., Wood Pulp, [1993] ECR 1307, paragraph 63.

42 However, where only one undertaking discloses information and (an)other undertaking(s) accept(s) it, there can also be a concerted practice. See for example Joined Cases T-25/95 etc., Cimenteries CBR and Others v Commission, [2000] ECR II-491, paragraph 1849: "[…] the concept of concerted practice does in fact imply the existence of reciprocal contacts […]. That condition is met where one competitor discloses its future intentions or conduct on the market to another when the latter requests it or, at the very least, accepts it."

EN 17 EN

58. Information exchange is very often pro-competitive as it can lead to, for example, an intensification of competition or significant efficiency gains. However, in situations where the exchange of market information is liable to enable undertakings to be aware of market strategies of their competitors it may lead to restrictive effects on competition.43 The competitive outcome of information exchange depends on the characteristics of the market in which it takes place (such as concentration, transparency, stability, complexity etc.) as well as on the type of information that is exchanged, which may modify the relevant market environment towards one liable to coordination.

59. Communication of information among competitors may constitute an agreement, a concerted practice, or a decision of an association of undertakings with the object of fixing prices or sharing markets or customers. These types of information exchanges run the risk of being investigated and, ultimately, fined as cartels.44 Information exchange may also facilitate the implementation of a cartel by enabling companies to monitor whether the participants comply with the agreed terms. These types of exchanges of information will be assessed as part of the cartel.45

2.2. Assessment under Article 101(1)

2.2.1. Main competition concerns

60. Conditional on the fact that the presence of an agreement, a concerted practice or a decision of an association of undertakings has been established, the following paragraphs set out the main competition concerns pertaining to information exchanges.

Collusive outcome

61. By increasing transparency in the market, the exchange of commercially sensitive (i.e., strategically useful) information can lead to the coordination (i.e., alignment) of companies' competitive behaviour and result in restrictive effects on competition. This can occur through different channels:

62. One way is that through information exchange companies may reach a common understanding on the terms of coordination, which can lead to a collusive outcome on the market. Information exchange can create mutually consistent expectations regarding the uncertainties present in the market. On this basis companies can then reach a common understanding on the terms of coordination of their competitive behaviour, even without an explicit agreement on coordination. Exchange of information about intentions on future conduct is the most likely to enable companies to reach such a common understanding. However, information on current conduct that reveals intentions on future behaviour can also be very useful in this context.

43 See Case C-7/95 P, John Deere, cited in note 18 above, paragraph 88 44 Note that in order to find that a concerted practice has an anti-competitive object, there does not need to

be a direct link between that practice and prices charged to final consumers; see Case C-8/08, T-Mobile Netherlands, cited in note 41 above, paragraph 39.

45 As set out in paragraph 9 of these guidelines, whereas these guidelines contain certain references to cartels, they are not intended to give any guidance as to what does and does not constitute a cartel as defined by the decisional practice of the Commission and the jurisprudence of the General Court and the Court of Justice of the European Union.

EN 18 EN

63. Another channel through which information exchange can lead to restrictive effects on competition is by increasing the internal stability of a collusive outcome on the market. In particular, it can do so by enabling the companies involved to monitor deviations. Namely, information exchange can make the market sufficiently transparent to allow the colluding companies to monitor to a sufficient degree whether other companies are deviating from the collusive outcome, and thus to know when to retaliate. Both exchanges of present and past data can constitute such a monitoring mechanism. This can either enable companies to reach a collusive outcome on markets where they were otherwise not able to do so, or it can increase stability of a collusive outcome already present on the market (see Example 3). However, as explained above, when a monitoring mechanism is part of a cartel, it will be assessed as part of that cartel and be treated as a restriction of competition by object.

64. A third channel through which information exchange can lead to restrictive effects on competition is by increasing the external stability of a collusive outcome on the market. Information exchanges that make the market sufficiently transparent can allow colluding companies to monitor where and when other companies are attempting to enter the market, thus allowing the colluding companies to target the new entrant. This may also tie into the anticompetitive foreclosure concerns below. Both exchanges of present and past data can constitute such a monitoring mechanism.

Anticompetitive foreclosure

65. An exclusive exchange of information could lead to anti-competitive foreclosure on the same market where the exchange takes place. This can occur when the exchange of commercially sensitive information places unaffiliated competitors at a significant competitive disadvantage as compared to the companies affiliated within the exchange system. This type of foreclosure is only possible if the information concerned is very strategic for competition and covers a significant part of the relevant market.

66. It cannot be excluded that information exchange may also lead to anticompetitive foreclosure of third parties in a related market. For instance, by gaining enough market power, parties exchanging information in an upstream market may be able to raise the price of a key component for a market downstream. Thereby, they could raise the costs of their rivals downstream, which could result in anti-competitive foreclosure.

2.2.2. Restriction of competition by object

67. Exchanging information on intentions of future conduct regarding prices or quantities46 is particularly likely to lead to a collusive outcome. Informing each other about such intentions may allow competitors to arrive at a common higher price level without incurring the risk of losing market share or triggering a price war during the period of adjustment to new prices. Moreover, it is less likely that this type of information exchange is done for pro-competitive reasons.

46 Information regarding intended future quantities could for instance include intended future sales, market

shares, territories, or customer lists.

EN 19 EN

68. Information exchanges between competitors of individualised data regarding intended future prices or quantities should therefore be considered a restriction of competition by object within the meaning of Article 101(1)47. This also applies to information exchanges on current conduct that reveals intentions on future behaviour and to cases where the combination of different types of data enables the direct deduction of intended future prices or quantities. This is without prejudice to the fact that there may be other types of information exchanges (mainly private individualised exchanges between competitors on prices and market shares) whose aim is to restrict competition on the market, which would normally be considered as restrictive by object. These types of information exchanges run the risk of being investigated and, ultimately, fined as cartels48.

2.2.3. Restrictive effects on competition

69. The likely effects of an information exchange on competition (except for the exchange of individualised data on intended future prices or quantities, or other exchanges, which are considered as restrictions of competition by object) must be analysed on a case-by-case basis as the results of the assessment depend on a combination of case specific factors. The assessment must compare the likely effects of the information exchange with the competitive situation that would prevail in the absence of the information exchange49. For an information exchange to have restrictive effects on competition within the meaning of Article 101(1), it must be likely to have an appreciable adverse impact on one (or several) of the parameters of competition such as price, output, product quality, product variety or innovation. Whether or not an exchange of information will have restrictive effects depends on both the economic conditions on the relevant markets and the characteristics of information exchanged.

70. Certain market conditions may make coordination easier to reach, sustain internally, or sustain externally. Exchanges of information in such markets may have more restrictive effects compared to markets with different conditions. However, even where market conditions are such that coordination may be difficult to sustain before the exchange, the exchange of information may change the market conditions in a way that coordination becomes possible after the exchange - for example by increasing transparency in the market, reducing market complexity, buffering instability or compensating for asymmetry. For this reason it is important to assess the restrictive effects of the information exchange in the context of both the initial market conditions, and how the information exchange changes these conditions. This will include an assessment of the specific characteristics of the system concerned in particular, its purpose and the conditions of access to it and participation in it, as well as the type of information exchanged – be that, for example, public or confidential, aggregated or detailed, historical or current –, the periodicity of the exchange of information and its importance for the fixing of prices, volumes or conditions of service50. The following factors are relevant for this assessment.

47 The notion of "intended future prices" is illustrated in Example 1. 48 Note that in order to find that a concerted practice has an anti-competitive object, there does not need to

be a direct link between that practice and prices charged to final consumers; see Case C-8/08, T-Mobile Netherlands, cited in note 45 above, paragraph 39.

49 Case C-7/95 P, John Deere v Commission, cited in note 18 above, paragraph 76. 50 Case C-238/05, Asnef-Equifax, cited in note 18 above, paragraph 54.

EN 20 EN

(i) Market coverage

71. For an information exchange to be likely to have appreciable restrictive effects on competition, the companies involved in the exchange have to cover a sufficiently large part of the relevant market. Otherwise, the competitors that are not participating in the information exchange could constrain any anticompetitive behaviour of the companies involved. For example, by pricing below the coordinated price level companies unaffiliated within the information exchange system could threaten the external stability of a collusive outcome.

72. What constitutes "a sufficiently large part of the market" cannot be defined in the abstract and will depend on the specific facts of each case and the type of information exchange in question. Where, however, an information exchange takes place in the context of another type of horizontal co-operation agreement and does not go beyond what is necessary for its implementation, market coverage below the market share thresholds set out in the relevant chapter of these guidelines, the Block Exemption Regulation or the De Minimis Notice51 pertaining to the type of agreement in question will usually not be large enough for the information exchange to give rise to restrictive effects on competition.

(ii) Market characteristics

73. Companies are more likely to reach a collusive outcome in markets which are sufficiently transparent, concentrated, simple, stable and symmetric. However, information exchange can also enable companies to reach a collusive outcome in other market situations where they would not be able to do so in the absence of the information exchange. Information exchange can thereby facilitate a collusive outcome by increasing transparency in the market, reducing market complexity, buffering instability or compensating for asymmetry. In this context, the competitive outcome of an information exchange depends not only on the initial characteristics of the market in which it takes place (such as concentration, transparency, stability, complexity etc.), but also on how the type of the information exchanged may change those characteristics.52

74. Collusive outcomes are more likely in transparent markets. Transparency can facilitate collusion by enabling companies to reach a common understanding on the terms of coordination, or/and by increasing internal and external stability of collusion. Information exchange can increase transparency and hence limit uncertainties about the strategic variables of competition (e.g., prices, output, demand, costs etc.). The lower the pre-existing level of transparency in the market, the more value an information exchange may have in attaining a collusive outcome. However, an information exchange that contributes little to the transparency in a market is less likely to have appreciable negative effects than an information exchange that significantly increases transparency. Therefore it is the combination of both the pre-existing level of transparency and how the information exchange changes this level that will determine how likely the information will have negative

51 Cited in note 31 above. 52 It should be noted that the discussion below is not a complete list of relevant characteristics. There may

be other characteristics of the market which are important in the setting of certain information exchanges.

EN 21 EN

appreciable effects. The pre–existing degree of transparency, inter alia, depends on the number of market participants and the nature of transactions, which can range from public transactions to confidential bilateral negotiations between buyers and sellers. When evaluating the change in the level of transparency in the market, the key element is to identify to what extent the available information can be used by companies to determine the actions of their competitors.

75. Tight oligopolies can facilitate a collusive outcome on the market as it is easier for fewer companies to reach a common understanding on the terms of coordination and to monitor deviations. A collusive outcome is also more likely to be sustainable with fewer companies. With more companies coordinating, the gains from deviating are greater because a larger market share can be gained through undercutting. At the same time, gains from the collusive outcome are smaller because when there are more companies the share in the rents from the collusive outcome declines. Exchanges of information in such markets are more likely to cause appreciable negative effects as by increasing transparency, or modifying the market environment in another way (see below), they may facilitate coordination and monitoring among more companies than would be possible in its absence.

76. Companies may find it difficult to reach a collusive outcome in a complex market environment. However the use of information exchange may simplify such environments. In a complex market environment more information exchange is normally needed to reach a common understanding on the terms of coordination and to monitor deviations. For example, it may be easier to achieve a collusive outcome on a price for a single, homogeneous product, than on hundreds of prices in a market with many differentiated products. It is nonetheless possible that to circumvent the difficulties involved in reaching a collusive outcome on a large number of prices, companies may exchange information to establish simple pricing rules (e.g., pricing points).

77. Collusive outcomes are more likely where the demand and supply conditions are relatively stable.53 In an unstable environment it may be difficult for a company to know whether its lost sales are due to an overall low level of demand or due to a competitor offering particularly low prices, and therefore it is difficult to sustain a collusive outcome. In this context, volatile demand, substantial internal growth by some companies in the market, or frequent entry by new companies, may indicate that the current situation is not sufficiently stable for coordination to be likely.54 Information exchange in certain situations can serve the purpose of increasing stability in the market, and thereby may enable a collusive outcome in the market. Moreover, in markets where innovation is important, coordination may be more difficult since particularly significant innovations may allow one company to gain a major advantage over its rivals. For a collusive outcome to be sustainable, the reactions of outsiders, such as current and future competitors not participating in the coordination, as well as customers, should not be capable of jeopardising the results expected from the collusive outcome. In this context, the existence of barriers to

53 See Case T-35/92, John Deere v Commission, [1994] ECR II-957, paragraph 78. 54 See Commission Decision No 92/157/EEC in Case IV/31.370, UK Agricultural Tractor Registration

Exchange, OJ L68, 13.3.1992, p. 19, paragraph 51 and Case T-35/92, John Deere v Commission, cited in note 53 above, paragraph 78. Note that it is not necessary that absolute stability be established or fierce competition excluded.

EN 22 EN

entry makes it more likely that a collusive outcome on the market is feasible and sustainable.

78. A collusive outcome is more likely in symmetric market structures. When companies are homogenous in terms of their costs, demand, market shares, product range, capacities etc., they are more likely to reach a common understanding on the terms of coordination. However, information exchange may also allow a collusive outcome to occur in more heterogeneous market structures. Information exchange could make companies aware of their differences and help them to design means to accommodate for their heterogeneity in the context of coordination.

79. The stability of a collusive outcome also depends on the companies' discounting of future profits. The more companies value the current profits that they could gain from undercutting versus all the future ones that they could gain by the collusive outcome, the more unlikely it is that they will be able to reach a collusive outcome.

80. A collusive outcome is more likely among companies that will continue to operate in the same market for a long time as there they will be more committed to coordinate. If a company knows that it will interact with the others for a long time, it will have more incentives to reach the collusive outcome because the stream of future profits from the collusive outcome will be worth more than the short term profit it could have if it deviated.

(iii) Characteristics of the information exchange

Commercially sensitive information

81. The exchange of commercially sensitive, i.e. strategically useful data, between competitors is more likely to be caught by Article 101 than exchanges of other types of information. Sharing of strategic data can give rise to restrictive effects on competition if it reduces the parties' decision-making independence by decreasing their incentives to compete. Strategic information can be related to prices (actual prices, discounts, increases, reductions, rebates), customer lists, production costs, quantities, turnovers, sales, capacities, qualities, marketing plans, risks, plans, investments, technologies, R&D programs and results. Generally, information related to prices and quantities is the most strategic, followed by information about costs and demand. However, if companies compete with regard to R&D it is the technology data that may be the most strategic for competition. The strategic usefulness of the data also depends on the way it is presented and its age.

Public/non-public data

82. In general, exchanges of genuinely public information are unlikely to constitute an infringement of Article 101.55Genuinely public information is information that is equally easy (i.e., costless) to access for everyone. Even if the data is in what is often referred to as 'the public domain', it is not genuinely public if the costs involved in collecting the data discourage to a sufficient degree other companies and buyers from doing so. For information to be genuinely public, obtaining it should not be more

55 Joined Cases T-191/98, T-212/98 and T-214/98, Atlantic Container Line (TACA), [2003] ECR II-3275,

paragraph 1154. Note that this may not be the case if the exchange underpins a cartel.

EN 23 EN

costly for buyers and companies unaffiliated to the exchange system than for the companies exchanging the information. A possibility to gather the information in the market, for example to collect it from customers, does not necessarily mean that such information constitutes market data readily accessible to competitors.56

83. Even if there is public availability of data (e.g., information published by regulators), the existence of an additional information exchange by competitors may restrict competition. In this case, it is the incremental information that is critical to tip the market balance towards a collusive outcome.

Public/non-public exchange of information

84. An information exchange is genuinely public if it makes the exchanged data equally accessible to all competitors and buyers. The fact that information is exchanged in public may decrease the likelihood of a collusive outcome on the market to the extent that competitors unaffiliated to the information exchange, potential entrants, and buyers may be able to constrain potential restrictive effect on competition.

Aggregated/individualised data

85. Exchanges of genuinely aggregated data, i.e., where the recognition of individualised company level information is sufficiently difficult, are less likely to lead to restrictive effects on competition than exchanges of company level data. The exchange of individualised data facilitates a common understanding on the market, and punishment strategies by allowing the coordinating companies to single out a deviator or entrant. Nevertheless, the exchange of aggregated data may also lead to a collusive outcome. For example, members of a tight and stable oligopoly exchanging aggregated data could automatically assume that someone defected from the collusive outcome when noticing a market price below a certain level (see Example 4).

Age of data

86. The exchange of historic data is unlikely to lead to a collusive outcome as it is unlikely to be indicative of the competitors' future conduct or provide a common understanding on the market.57 Moreover, the older the data, the less useful it would be for timely detection of deviations and thus as a credible threat of prompt retaliation. Whether data is genuinely historic depends on the specific characteristics of the relevant market such as the frequency of contract renewals (when they are indicative of the frequency of price re-negotiations). For example, data can be considered as historic if it is several times older than the average length of contracts in the industry (see Example 5).

Frequency of the information exchange

56 See Joined Cases T-202/98 etc., Tate & Lyle v Commission, [2001] ECR II-2035, paragraph 60. 57 For example, in past cases the Commission has considered the exchange of data which was more than

one year old as historic and as not restrictive of competition within the meaning of Article 101(1), whereas information less than one year old has been considered as recent; Commission Decision No 92/157/EEC in Case IV/31.370, UK Agricultural Tractor Registration Exchange, cited in note 54 above, paragraph 50; Commission Decision No 98/4/ECSC in Case IV/36.069, Wirtschaftsvereinigung Stahl, OJ L1, 3.1.1998, p. 10, paragraph 17.

EN 24 EN

87. Frequent exchanges of information that facilitate both a better common understanding of the market and monitoring of deviations increase the risks of a collusive outcome. In markets with long term contracts (which are indicative of infrequent price re-negotiations) a less frequent exchange of information could therefore normally be sufficient for reaching a collusive outcome than in markets with short term contracts and frequent price re-negotiations. However, depending on the structure of the market, it is possible that an isolated exchange may constitute a sufficient basis for the participating undertakings to concert their market conduct (i.e., reach a common understanding on the terms of coordination) and thus successfully substitute practical cooperation between them for competition and the risks that that entails.58


2.3.1. Efficiency gains

88. Information exchange may lead to significant efficiency gains. Information about competitors' costs can enable companies to become more efficient if they benchmark their performance against the best practices in the industry and design internal incentive schemes accordingly.

89. Moreover, in certain situations information exchange can help companies allocate production towards high-demand markets (e.g., demand information) or low cost companies (e.g., cost information). The likelihood of these types of efficiencies depends on market characteristics (nature of uncertainties, strategic variable of competition, etc.). Some forms of information exchanges in this context may allow substantial cost savings where for example they reduce unnecessary inventories, or enable quicker delivery of perishable products to areas with high demand and their reduction in areas with low demand (see Example 7).

90. Exchange of consumer data between companies in markets with asymmetric information about consumers can also give rise to efficiencies. For instance, keeping track of the past behaviour of customers in terms of accidents or credit default provides an incentive for consumers to limit their risk exposure. It also allows detecting which consumers carry a lower risk and should benefit from lower prices. In this context, information exchange can also reduce consumer lock-in, thereby inducing stronger competition. This is because information is generally specific to a relationship and consumers would otherwise lose the benefit from this information when switching to another company. Examples of such efficiencies are found in the banking and insurance sectors, which are characterised by frequent exchanges of information about consumer defaults and risk characteristics.

91. Exchanging past and present market share information may provide benefits to both companies and consumers by allowing companies to announce it as a signal of quality to consumers. In situations of imperfect information about product quality, consumers often use indirect means to gain information on the relative qualities of products such as price and market shares (e.g., consumers use best-selling lists in order to choose their next book). Especially in markets with switching costs or network externalities, announcing market shares provides a strong (credible)

58 See Case C-8/08, T-Mobile Netherlands, cited in note 41 above, paragraph 59.

EN 25 EN

indicator of future performance. In this context an information exchange about sales can benefit both companies and consumers.

92. Information exchange that is genuinely public can also benefit consumers by helping them to make a more informed choice (by reducing their search costs). However, for such direct consumer benefits to arise, companies also need to be committed to announce the true values of their intentions. This can happen if companies have repeated interactions with consumers.59 Similarly, information exchange about input prices can lower search costs for companies that could ultimately benefit consumers.

93. As shown above, exchanging present and past data is more likely to generate efficiency gains than exchanging information about future intentions. However, in specific circumstances announcing future intentions could also give rise to efficiency gains. For example, companies knowing early the winner of an R&D race could avoid wasting resources.

2.3.2. Indispensability

94. Restrictions that go beyond what is necessary to achieve the efficiency gains generated by an information exchange do not fulfil the conditions of Article 101(3). For fulfilling the condition of indispensability, the parties will need to prove that the data's type, aggregation, age, confidentiality and frequency of the exchange are of the type that carries the lowest risks indispensable for creating the claimed efficiency gains. Moreover, the exchange should not involve information beyond the variables that are relevant for the attainment of the efficiency gains. For instance, for the purpose of benchmarking, an exchange of individualised data would generally not be indispensable because information aggregated in for example some form of industry ranking could also generate the claimed efficiency gains while carrying a lower risk of leading to a collusive outcome (see Example 4). Finally, it is generally unlikely that the sharing of future data is indispensable, especially if it is related to prices and quantities.

95. Similarly, information exchanges that form part of horizontal co-operation agreements are also more likely to fulfil the conditions of Article 101(3) if they do not go beyond what is indispensable for the implementation of the economic purpose of the agreement (e.g., sharing technology necessary for an R&D agreement or cost data in the context of a production agreement).

2.3.3. Pass-on to consumers

96. Efficiency gains attained by indispensible restrictions must be passed on to consumers to an extent that outweighs the restrictive effects on competition caused by an information exchange. The lower is the market power of the parties involved in the information exchange, the more likely it is that the efficiency gains would be passed on to consumers to an extent that outweighs the restrictive effects on competition.

59 However, it should not be presumed that a price commitment vis-à-vis consumers is always good,

which is demonstrated by practices such as price matching or price beating guarantees as these could deter competitors from lowering their price.

EN 26 EN

2.3.4. No elimination of competition

97. The criteria of Article 101(3) cannot be met if the companies involved in the information exchange are afforded the possibility of eliminating competition in respect of a substantial part of the products concerned.

2.4. Examples

98. Exchange of intended future prices as a restriction by object

Example 1 Situation: A trade association of coach companies in country X disseminates individualised information on intended future prices to the member coach companies. The information contains several elements, such as the intended fare and the route to which the fare applies, the possible restrictions to this fare, such as which consumers can buy it, if advanced payment or minimum stay is required, the period during which tickets can be sold for the given fare (first and last ticket date), and the time during which the ticket with the given fare can be used for travel (first and last travel dates). Analysis: This information exchange, which is triggered by a decision of an association of undertakings, restricts competition by object within the meaning of Article 101(1). Moreover, it is likely to be investigated, and, ultimately, fined as a cartel. In this case, the exchange of information concerns pricing intentions of competitors. This information exchange is a very efficient tool for reaching a collusive outcome and therefore restricts competition by object within the meaning of Article 101(1). This is because the companies are free to change the announced prices at any time if they learn that their competitors intend to charge higher prices. This allows the companies to reach a common higher price level without incurring the cost of losing market share. For example, coach company A can announce today a price increase on the route from city 1 to city 2 for travel as of the following month. Since this information is accessible to all other coach companies, company A can then wait and see the reaction of its competitors to this price announcement. If a competitor on the same route, say company B, matched the price increase, then A's announcement would be left unchanged and later would likely become effective. However, if company B did not match the price increase, then company A could still revise its fare. The adjustment would continue until the companies converged to an increased anticompetitive price level. This information exchange is unlikely to fulfil the conditions of Article 101(3). The information exchange is only confined to competitors, i.e., customers of the coach companies do not directly benefit from it.

99. Exchange of current prices with sufficient efficiency gains for consumers

Example 2 Situation: A national tourist office together with the coach companies in country X agree to disseminate information on current prices of coach tickets through a freely accessible website (in contrast to the example above consumers can already purchase tickets at the prices and conditions which are exchanged, thus they are not intended future prices but present prices of current and future services). The information contains several elements, such as the fare and the route to which the

EN 27 EN

fare is applied, the possible restrictions to this fare, such as which consumers can buy it, if advanced payment or minimum stay is required, and the time during which the ticket with the given fare can be used for travel (first and last travel dates). Coach travel in country X is not in the same relevant market as train and air travel. The relevant market is concentrated, stable and relatively non-complex. Analysis: This information exchange does not constitute a restriction of competition by object. The companies are exchanging current prices rather than intended future prices because they are already selling tickets at these prices (unlike in Example 1). Therefore, this exchange of information is not likely to constitute an efficient mechanism for reaching a focal point for coordination. Nevertheless, given the market structure and strategic nature of the data, this information exchange is likely to constitute an efficient mechanism for monitoring deviations from a collusive outcome. Therefore, this information exchange could give rise to restrictive effects on competition within the meaning of Article 101(1). However, to the extent that some restrictive effects on competition could result from the possibility to monitor deviations, it is likely that the efficiency gains stemming from the information exchange would be passed on to consumers to an extent that outweighs the likely restrictive effects on competition. Unlike in Example 1, the information exchange is public and consumers can actually purchase tickets at prices and conditions which are exchanged. Therefore this information exchange is likely to directly benefit consumers by reducing their search costs and improving choice, and thereby also stimulating price competition. Hence, the conditions of Article 101(3) are likely to be met.

100. Current prices deduced from the information exchanged

Example 3

Situation: Luxury hotels in the capital of country A, which is a tight and stable oligopoly operating in a non complex and concentrated market, directly exchange individual information about current occupancy rates and revenues. In this case, from the information exchanged the parties can directly deduce their actual current prices.

Analysis: Unless it reveals intentions of future behaviour, this exchange of information does not in principle60 constitute a restriction of competition by object because the hotels exchange present data and not information on intended future prices or quantities. However, the information exchange would give rise to restrictive effects on competition within the meaning of Article 101(1) because knowing the competitor's actual current prices would be likely to lead to coordination (i.e., alignment) of companies' competitive behaviour. It would be most likely used to monitor deviations from the collusive outcome. The information exchange increases transparency in the market as even though the hotels normally publish their list prices, they also offer various discounts to the list price resulting from negotiations or for early or group bookings, etc. Therefore, the incremental information that is non-publicly exchanged between the hotels is commercially sensitive, i.e., strategically useful. This exchange is likely to lead to a collusive

60 This is without prejudice to the fact that if this information exchange was found to aim at restricting

competition, it would normally constitute a restriction of competition by object.

EN 28 EN

outcome because the parties involved constitute a tight, non complex and stable oligopoly involved in a long term competitive relationship (repeated interactions). Moreover, the cost structures of the hotels are largely homogeneous. Finally, neither consumers nor entry can constrain the incumbents' anti-competitive behaviour as consumers have little buyer power and barriers to entry are high. It is unlikely that in this case the parties would be able to demonstrate any efficiency gains stemming from the information exchange that would be passed on to consumers to an extent that would outweigh the restrictive effects on competition. Therefore it is unlikely that the conditions of Article 101(3) can be met.

101. Benchmarking benefits – criteria of Article 101(3) not fulfilled

Example 4 Situation: Three large companies with a combined market share of 80% in a stable, non-complex, concentrated market with high barriers to entry non-publicly and frequently exchange information directly between themselves about their individual costs. The companies do this to benchmark their performance against their competitors and thereby intend to become more efficient. Analysis: This information exchange does not in principle constitute a restriction of competition by object. Consequently, its effects on the market need to be assessed. Because of the market structure, the individualised form of presentation of the data and its large coverage of the relevant market, the information exchange is likely to lead to a collusive outcome and thereby gives rise to restrictive effects on competition within the meaning of Article 101(1). It is unlikely that the criteria of Article 101(3) are fulfilled because there are less restrictive means to achieve the claimed efficiency gains, for example by aggregating the data in some form of industry ranking. Only the information that is relevant to the identification of the best performing practice should have been released in this context. Individualised data in this case could actually be counter-productive as it would make the relative incentive schemes sensitive to the company specific qualities and shocks. Finally, in this case, since the parties form a tight, non-complex and stable oligopoly, even the exchange of aggregated data may lead to a collusive outcome if it allows the parties to set a focal point for coordination.

102. Historic data

Example 5 Situation: Companies are in an industry characterised by short term contracts and where prices are re-negotiated every three months directly exchange price data that is three years old. Analysis: In a context of contract renewal and price re-negotiations taking place every three months, three year old data constitute historic information and their exchange would not be likely to lead to a collusive outcome in the market and thereby give rise to restrictive effects on competition within the meaning of Article 101(1).

103. Genuinely public information

Example 6

EN 29 EN

Situation: The four companies owning all the petrol stations in country A exchange current gasoline prices over the telephone. They claim that this information exchange cannot have restrictive effects on competition because the information is public as it is displayed on large display panels at every petrol station. Analysis: The pricing data exchanged over the telephone is not genuinely public as it provides an informational advantage by means of cost savings to the participating companies. Consumers and entrants that want to obtain the same information would need to incur substantial time and transport costs because they would have to collect the information themselves by going to the different petrol stations to look at the prices displayed on the boards. The costs for this are potentially so high that the information could not be obtained but for the information exchange. Moreover, the exchange is systematic and covers the entire relevant market, which is a tight, non-complex, stable oligopoly. Therefore it is likely to create a climate of mutual certainty as to the competitors' pricing policy and thereby it is likely to lead to a collusive outcome. Consequently, this information exchange is likely to give rise to restrictive effects on competition within the meaning of Article 101(1).

104. Improved meeting of demand as an efficiency gain

Example 7 Situation: There are five producers of fresh bottled carrot juice in the relevant market. Demand for this product is very unstable and heterogeneous. The juice has to be sold and consumed within one day from the date of production. The producers agree to establish an independent market research company that on a daily basis collects information about unsold juice in each output point, which it publishes on its website the following week in a form that is aggregated per output point. The published statistics allow producers and retailers to forecast demand and to better position the product. The producers demonstrate a clear market failure in the absence of the information exchange system. Namely, before the exchange was put in place, the retailers reported large quantities of wasted juice and therefore reduced the quantity of juice purchased from the producers. Consequently, because of the retailers' risk aversion, in some periods and areas there were frequent instances of unmet demand. The information exchange system, which allowed better forecasting of oversupply and undersupply, significantly reduced the instances of unmet consumer demand and increased the demand for the juice by retailers.

Analysis: To the extent that this information exchange gives rise to restrictive effects on competition within the meaning of Article 101(1), due to the concentrated market structure and the strategic nature of the information, they are likely to be outweighed by efficiency gains stemming from increasing supply to places with high demand and decreasing supply in places with low demand. The information is exchanged in a public and aggregated form which carries lower anti-competitive risks than if it was non-public and individualised. The information exchange therefore does not go beyond what is necessary to correct the market failure which has been established by the parties. Therefore, it is likely that this information exchange meets the criteria of Article 101(3).

EN 30 EN

3. RESEARCH AND DEVELOPMENT AGREEMENTS

3.1. Definition

105. R&D agreements vary in form and scope. They range from outsourcing certain R&D activities to the joint improvement of existing technologies and to a co-operation concerning the research, development and marketing of completely new products. They may take the form of a co-operation agreement or of a jointly controlled company. This chapter applies to all forms of R&D agreements, including related agreements concerning the production or commercialisation of the R&D results provided that the R&D agreement is the most upstream indispensible building block of the co-operation.

3.2. Relevant markets

106. The key to defining the relevant market when assessing the effects of an R&D agreement is to identify those products, technologies or R&D efforts, that will act as the main competitive constraints on the parties. At one end of the spectrum of possible situations, the innovation may result in a product (or technology) which competes in an existing product (or technology) market. This is, for example, the case with R&D directed towards slight improvements or variations, such as new models of certain products. Here possible effects concern the market for existing products. At the other end of the spectrum, innovation may result in an entirely new product which creates its own new product market (e.g., a new vaccine for a previously incurable disease). However, many cases concern situations in between these two extremes, i.e., situations in which innovation efforts may create products (or technology) which, over time, replace existing ones (e.g., CDs which have replaced records). A careful analysis of those situations may have to cover both existing markets and the impact of the agreement on innovation.

Existing product markets

107. When the co-operation concerns R&D for the improvement of existing products, these existing products and their close substitutes form the relevant market concerned by the co-operation.61

108. If the R&D efforts aim at a significant change of existing products or even at a new product replacing existing ones, substitution with the existing products may be imperfect or long-term. It may be concluded that the old and the potentially emerging new products do not belong to the same relevant market.62 The market for existing products may nevertheless be concerned, if the pooling of R&D efforts is likely to result in the coordination of the parties’ behaviour as suppliers of existing products, for instance because of the exchange of competitively sensitive information relating to the market for existing products.

61 For market definition, see the Market Definition Notice, cited in note 5 above. 62 See also Commission Guidelines on the application of Article 81 of the EC Treaty to technology

transfer agreements, OJ C101, 27.4.2004, p. 2, paragraph 23 ("Technology Transfer Guidelines"), paragraph 33.

EN 31 EN

109. If the R&D concerns an important component of a final product, not only the market for this component may be relevant for the assessment, but the existing market for the final product as well. For instance, if car manufacturers co-operate in R&D related to a new type of engine, the car market may be affected by this R&D co-operation. The market for final products, however, is only relevant for the assessment if the component at which the R&D is aimed is technically or economically a key element of these final products and if the parties to the R&D agreement have market power with respect to the final products.

Existing technology markets

110. R&D co-operation may not only concern products but also technology. When rights to intellectual property are marketed separately from the products to which they relate, the relevant technology market has to be defined as well. Technology markets consist of the intellectual property that is licensed and its close substitutes, i.e., other technologies which customers could use as a substitute.

111. The methodology for defining technology markets follows the same principles as product market definition.63 Starting from the technology which is marketed by the parties, those other technologies to which customers could switch in response to a small but non-transitory increase in relative prices need to be identified. Once these technologies are identified, market shares can be calculated by dividing the licensing income generated by the parties with the total licensing income of all licensors.

112. The parties’ position in the market for existing technology is a relevant assessment criterion where the R&D co-operation concerns a significant improvement of an existing technology or a new technology that is likely to replace the existing technology. The parties’ market shares can, however, only be taken as a starting point for this analysis. In technology markets, particular emphasis must be put on potential competition. If companies, who do not currently license their technology, are potential entrants on the technology market they could constrain the ability of the parties to profitably raise the price for their technology. This aspect of the analysis may also be taken into account directly in the calculation of market shares by basing these on the sales of the products incorporating the licensed technology on downstream product markets (see the section on calculation of market shares below).

Competition in innovation (R&D efforts)

113. R&D co-operation may not – or not only – affect competition in existing markets, but also competition in innovation and new product markets. This is the case where R&D co-operation concerns the development of new products/technology which either may – if emerging – one day replace existing ones or which are being developed for a new intended use and will therefore not replace existing products but create a completely new demand. The effects on competition in innovation are important in these situations, but can in some cases not be sufficiently assessed by analysing actual or potential competition in existing product/technology markets. In this respect, two scenarios can be distinguished, depending on the nature of the innovative process in a given industry.

63 See Market Definition Notice, cited in note 5 above; see also Technology Transfer Guidelines, cited in

note 62 above, paragraphs 19 et seqq.

EN 32 EN

114. In the first scenario, which is, for instance, present in the pharmaceutical industry, the process of innovation is structured in such a way that it is possible at an early stage to identify competing R&D poles. Competing R&D poles are R&D efforts directed towards a certain new product or technology, and the substitutes for that R&D, i.e. R&D aimed at developing substitutable products or technology for those developed by the co-operation and having similar timing. In this case, it can be analysed if after the agreement there will be a sufficient number of remaining R&D poles. The starting point of the analysis is the R&D of the parties. Then credible competing R&D poles have to be identified. In order to assess the credibility of competing poles, the following aspects have to be taken into account: the nature, scope and size of possible other R&D efforts, their access to financial and human resources, know how/patents, or other specialised assets as well as their timing and their capability to exploit possible results. An R&D pole is not a credible competitor if it cannot be regarded as a close substitute for the parties’ R&D effort from the viewpoint of, for instance, access to resources or timing.

115. Besides the direct effect on the innovation itself, the co-operation may also affect a new product market. It will often be difficult to analyse the effects on such a market directly as by its very nature it does not yet exist. The analysis of such markets will therefore often be implicitly incorporated in the analysis of competition in innovation. However, it may be necessary to consider directly the effects on such a market of aspects of the agreement that go beyond the R&D stage. An R&D agreement that includes joint production and commercialisation on the new product market may, for instance, be assessed differently than a pure R&D agreement.

116. In the second scenario, the innovative efforts in an industry are not clearly structured so as to allow the identification of R&D poles. In this situation, the Commission would, absent exceptional circumstances, not try to assess the impact of a given R&D co-operation on innovation, but would limit its assessment to existing product and/or technology markets which are related to the R&D co-operation in question.

Calculation of market shares

117. The calculation of market shares, both for the purposes of the R&D Block Exemption Regulation64 and of these guidelines, has to reflect the distinction between existing markets and competition in innovation. At the beginning of an R&D co-operation the reference point is the existing market for products capable of being improved or replaced by the products under development. If the R&D agreement only aims at improving or refining existing products, this market includes the products directly concerned by the R&D. Market shares can thus be calculated on the basis of the sales value of the existing products.

118. If the R&D aims at replacing an existing product, the new product will, if successful, become a substitute to the existing products. To assess the competitive position of the parties, it is again possible to calculate market shares on the basis of the sales value of the existing products. Consequently, the R&D Block Exemption Regulation bases its exemption of these situations on the market share in “the relevant market for the products capable of being improved or replaced by the contract products”. To

64 Cited in note 3 above.

EN 33 EN

fall under the R&D Block Exemption Regulation, this market share may not exceed 25%.65

119. For technology markets one way to proceed is to calculate market shares on the basis of each technology's share of total licensing income from royalties, representing a technology's share of the market where competing technologies are licensed. However, this may often be a mere theoretical and not very practical way to proceed because of lack of clear information on royalties, the use of royalty free cross-licensing, etc. An alternative approach is to calculate market shares on the technology market on the basis of sales of products or services incorporating the licensed technology on downstream product markets. Under this approach all sales on the relevant product market are taken into account, irrespective of whether the product incorporates a technology that is being licensed.66 Also for this market the share may not exceed 25% (irrespective of the calculation method used) for the benefits of the R&D Block Exemption Regulation to apply.

120. If the R&D aims at developing a product which will create a completely new demand, market shares based on sales cannot be calculated. Only an analysis of the effects of the agreement on competition in innovation is possible. Consequently, the R&D Block Exemption Regulation treats these agreements as agreements between non-competitors and exempts them irrespective of market share for a period of seven years after the product is first put on the market.67 However, the benefit of the block exemption may be withdrawn if the agreement eliminated effective competition in innovation.68 After the seven year period, market shares based on sales value can be calculated, and the market share threshold of 25% applies.69



121. R&D co-operation can restrict competition in various ways. First, it may reduce or slow down innovation, leading to fewer or worse products coming to the market later than they otherwise would. Secondly, on goods or technology markets the R&D co-operation may reduce significantly competition between the parties outside the scope of the agreement or it may make anti-competitive coordination on these markets likely. A foreclosure problem may only arise in the context of co-operation involving at least one player with a significant degree of market power (which does not necessarily amount to dominance) for a key technology and the exclusive exploitation of the results.

3.3.2. Restrictions of competition by object

122. R&D agreements constitute a restriction of competition by object if they do not truly concern joint R&D, but serve as a tool to engage in a disguised cartel, i.e., otherwise prohibited price fixing, output limitation or market allocation. However, an R&D

65 Article 4(2) of the R&D Block Exemption Regulation, cited in note 3 above. 66 See also Technology Transfer Guidelines, cited in note 62 above, paragraph 23. 67 Article 4(1) of the R&D Block Exemption Regulation, cited in note 3 above. 68 See recitals 19 and 20 of the R&D Block Exemption Regulation, cited in note 3 above. 69 Article 4(3) of the R&D Block Exemption Regulation, cited in note 3 above.

EN 34 EN

agreement which includes the joint exploitation of possible future results is not necessarily restrictive of competition.


123. Most R&D agreements do not fall under Article 101(1). First, this can be said for many agreements relating to co-operation in R&D at a rather early stage, far removed from the exploitation of possible results.

124. Moreover, R&D co-operation between non-competitors does generally not give rise to restrictive effects on competition.70 The competitive relationship between the parties has to be analysed in the context of affected existing markets and/or innovation. If the parties are not able to carry out the necessary R&D independently, the R&D agreement will normally not have any restrictive effects on competition. This can apply, for example, to companies bringing together complementary skills, technologies and other resources. The issue of potential competition has to be assessed on a realistic basis. For instance, parties cannot be defined as potential competitors simply because the co-operation enables them to carry out the R&D activities. The decisive question is whether each party independently has the necessary means as to assets, know how and other resources.

125. Outsourcing of previously captive R&D is a specific form of R&D co-operation. In such a scenario, the R&D is often carried out by specialised companies, research institutes or academic bodies which are not active in the exploitation of the results. Normally, such agreements are combined with a transfer of know how and/or an exclusive supply clause concerning the possible results, which, due to the complementary nature of the co-operating parties in these scenarios, do not give rise to restrictive effects on competition within the meaning of Article 101(1).

126. R&D co-operation which does not include the joint exploitation of possible results by means of licensing, production and/or marketing rarely gives rise to restrictive effects on competition within the meaning of Article 101(1). Those pure R&D agreements can only cause a competition problem if competition with respect to innovation is appreciably reduced, leaving only a limited number of credible competing R&D poles.

127. R&D agreements are only likely to give rise to restrictive effects on competition where the parties to the co-operation have market power on the existing markets and/or competition with respect to innovation is appreciably reduced.

128. There is no absolute threshold above which it can be presumed that an R&D agreement creates or maintains market power and thus is likely to give rise to restrictive effects on competition within the meaning of Article 101(1). However, R&D agreements between competitors are block exempted provided that their combined market share does not exceed 25% and that the other conditions for the application of the R&D Block Exemption Regulation are fulfilled.

70 An R&D co-operation between non-competitors can, however, produce foreclosure effects under

Article 101(1) if it relates to an exclusive exploitation of results and if it is concluded between companies, one of which has a significant degree of market power (which does not necessarily amount to dominance) with respect to a key technology.

EN 35 EN

129. Agreements falling outside the R&D Block Exemption Regulation because the combined market share of the parties exceeds 25% do not necessarily give rise to restrictive effects on competition. However, the stronger the combined position of the parties on existing markets and/or the more competition in innovation is restricted, the more likely it is that the R&D agreement can cause restrictive effects on competition.

130. If the R&D is directed at the improvement or refinement of existing products/technology, possible effects concern the relevant market(s) for these existing products/technology. Effects on prices, output, product quality, product variety or innovation in existing markets are, however, only likely if the parties together have a strong position, entry is difficult and few other innovation activities are identifiable. Furthermore, if the R&D only concerns a relatively minor input of a final product, effects as to competition in these final products are, if any, very limited.

131. In general, a distinction has to be made between pure R&D agreements and agreements foreseeing a more comprehensive co-operation involving different stages of the exploitation of results (i.e. licensing, production, marketing). As set out above, pure R&D agreements will only rarely give rise to restrictive effects on competition within the meaning of Article 101(1). This is in particular true for R&D directed towards a limited improvement of existing products/technology. If, in such a scenario, the R&D co-operation includes joint exploitation only by means of licensing to third parties, restrictive effects such as foreclosure problems are unlikely. If, however, joint production and/or marketing of the slightly improved products/technology are included, the effects on competition of the co-operation have to be examined more closely. Restrictive effects on competition in the form of increased prices or reduced output in existing markets are more likely if strong competitors are involved in such a situation.

132. If the R&D is directed at an entirely new product (or technology) which creates its own new market, price and output effects on existing markets are rather unlikely. The analysis has to focus on possible restrictions of innovation concerning, for instance, the quality and variety of possible future products/technology or the speed of innovation. Those restrictive effects can arise where two or more of the few companies engaged in the development of such a new product start to co-operate at a stage where they are each independently rather near to the launch of the product. Such effects are typically the direct result of the agreement between the parties. Innovation may be restricted even by a pure R&D agreement. In general, however, R&D co-operation concerning entirely new products is unlikely to give rise to restrictive effects on competition unless only a limited number of credible alternative R&D poles exist. This principle does not change significantly if the joint exploitation of the results, even joint marketing, is involved. In these situations the issue of joint exploitation may only give rise to restrictive effects on competition where foreclosure from key technologies plays a role. Those problems would, however, not arise where the parties grant licences that allow third parties to compete effectively.

133. Many R&D agreements will lie somewhere in between the two situations described above. They may therefore have effects on innovation as well as repercussions on existing markets. Consequently, both the existing market and the effect on innovation may be of relevance for the assessment with respect to the parties’ combined

EN 36 EN

positions, concentration ratios, number of players/innovators and entry conditions. In some cases there can be restrictive effects on competition in the form of increased prices or reduced output, innovation, product quality or variety in existing markets and in the form of a negative impact on innovation by means of slowing down the development. For instance, if significant competitors on an existing technology market co-operate to develop a new technology which may one day replace existing products this co-operation may slow down the development of the new technology if the parties have market power on the existing market and also a strong position with respect to R&D. A similar effect can occur if the major player in an existing market co-operates with a much smaller or even potential competitor who is just about to emerge with a new product/technology which may endanger the incumbent’s position.

134. Agreements may also fall outside the R&D Block Exemption Regulation irrespective of the parties' market power. This applies for instance to agreements which restrict equal access of a party to the results of the R&D co-operation.71 The R&D Block Exemption Regulation provides for a specific exception to this general rule in the case of academic bodies, research institutes or specialised companies which provide R&D as a service and which are not active in the industrial exploitation of the results of research and development.72 Nevertheless, agreements containing exclusive access rights may, where they fall under Article 101(1), fulfil the criteria of Article 101(3), particularly where exclusive access rights are economically indispensable in view of the market, risks and scale of the investment required to exploit the results of the research and development.



135. Many R&D agreements – with or without joint exploitation of possible results – bring about efficiency gains by means of combining complementary skills and assets, thus resulting in improved or new products and technologies being developed and marketed more rapidly than would otherwise be the case. R&D agreements may also lead to cost reductions.


136. Restrictions that go beyond what is necessary to achieve the efficiency gains generated by an R&D agreement do not fulfil the criteria of Article 101(3). The restrictions listed in Article 5 of the R&D Block Exemption Regulation may render it less likely to find that the criteria of Article 101(3) are met following an individual assessment. It will therefore generally be necessary for the parties to an R&D agreement to show that such restrictions are indispensable to the co-operation.


137. Efficiency gains attained by indispensible restrictions must be passed on to consumers to an extent that outweighs the restrictive effects on competition caused

71 See Art. 3(3) of the R&D Block Exemption Regulation, cited in note 3 above. 72 See Art. 3(3) of the R&D Block Exemption Regulation, cited in note 3 above.

EN 37 EN

by the R&D agreement. In general, it is more likely that an R&D agreement will bring about efficiency gains that benefit consumers if the R&D agreement results in the combination of complementary skills and assets. The parties to an agreement may, for instance, have different research capabilities. If, on the other hand, the parties' skills and assets are very similar, the most important effect of the R&D agreement may be the elimination of part or all of the R&D of one or more of the parties. This would eliminate (fixed) costs for the parties to the agreement but would be unlikely to lead to benefits which will be passed on to consumers. Moreover, the higher the market power of the parties the less likely they will pass on the efficiency gains to consumers to an extent that would outweigh the restrictive effects on competition.


138. The criteria of Article 101(3) cannot be met if the parties are afforded the possibility of eliminating competition in respect of a substantial part of the products (or technologies) in question.

3.4.5. Time of the assessment

139. The assessment of restrictive agreements under Article 101(3) is made within the actual context in which they occur and on the basis of the facts existing at any given point in time. The assessment is sensitive to material changes in the facts. The exception rule of Article 101(3) applies as long as the four conditions of Article 101(3) are fulfilled and ceases to apply when that is no longer the case. When applying Article 101(3) in accordance with these principles it is necessary to take into account the initial sunk investments made by any of the parties and the time needed and the restraints required to commit and recoup an efficiency enhancing investment. Article 101 cannot be applied without taking due account of such ex ante investment. The risk facing the parties and the sunk investment that must be committed to implement the agreement can thus lead to the agreement falling outside Article 101(1) or fulfilling the conditions of Article 101(3), as the case may be, for the period of time needed to recoup the investment. Should the invention resulting from the investment benefit from any form of exclusivity granted to the parties under rules specific to the protection of intellectual property rights, the recoupment period for such an investment will generally be unlikely to exceed the exclusivity period established under these rules.

140. In some cases the restrictive agreement is an irreversible event. Once the restrictive agreement has been implemented the ex ante situation cannot be re-established. In such cases the assessment must be made exclusively on the basis of the facts pertaining at the time of implementation. For instance, in the case of an R&D agreement whereby each party agrees to abandon its respective research project and pool its capabilities with those of another party, it may from an objective point of view be technically and economically impossible to revive a project once it has been abandoned. The assessment of the anti-competitive and pro-competitive effects of the agreement to abandon the individual research projects must therefore be made as of the time of the completion of its implementation. If at that point in time the agreement is compatible with Article 101, for instance because a sufficient number of third parties have competing R&D projects, the parties' agreement to abandon their individual projects remains compatible with Article 101, even if at a later point

EN 38 EN

in time the third party projects fail. However, the prohibition of Article 101 may apply to other parts of the agreement in respect of which the issue of irreversibility does not arise. If, for example, in addition to joint R&D, the agreement provides for joint exploitation, Article 101 may apply to this part of the agreement if due to subsequent market developments the agreement gives rise to restrictive effects on competition and does not (any longer) satisfy the conditions of Article 101(3) taking due account of ex ante sunk investments.

3.5. Examples

141. Impact of joint R&D on innovation markets/new product market

Example 1 Situation: A and B are the two major companies on the European market for the manufacture of existing electronic components. Both have a market share of 30%. They have each made significant investments in the R&D necessary to develop miniaturised electronic components and have developed early prototypes. They now agree to pool these R&D efforts by setting up a JV to complete the R&D and produce the components, which will be sold back to the parents, who will commercialise them separately. The remainder of the market consists of small companies without sufficient resources to undertake the necessary investments. Analysis: Miniaturised electronic components, while likely to compete with the existing components in some areas, are essentially a new technology and an analysis must be made of the poles of research destined towards this future market. If the JV goes ahead then only one route to the necessary manufacturing technology will exist, whereas it would appear likely that A and B could reach the market individually with separate products. The agreement therefore reduces product variety. The joint production is also likely to directly limit competition between the parties to the agreements and lead them to agree on output levels, quality or other competitively important parameters. This would limit competition even though the parties will commercialise the products independently. The parties could, for instance, limit the output of the JV compared to what the parties would have brought to the market if they had decided their output on their own. The JV could also charge a high transfer price to the parties, thereby increasing the input costs for the parties which could lead to higher downstream prices. The parties have a large combined market share of the parties on the existing downstream market and the remainder of this market is fragmented. This situation is likely to become even more pronounced on the new downstream product market since the smaller competitors cannot invest in the new components. It is therefore quite likely that the joint production will restrict competition. Furthermore, the market for miniaturised electronic components is in the future likely to develop into a duopoly with a high degree of commonality of costs and possible exchange of commercially sensitive information between the parties. There may therefore also be a serious risk of anti-competitive coordination leading to a collusive outcome in this market. The R&D agreement is therefore likely to give rise to restrictive effects on competition within the meaning of Article 101(1).While the agreement could give rise to efficiency gains in the form of bringing a new technology forward quicker, the parties would on the other hand face no competition at the R&D level, so their incentives to pursue the new technology at a high pace could be severely reduced.

EN 39 EN

Although some of these concerns could be remedied if the parties committed to license key know how for manufacturing miniature components to third parties on reasonable terms, it seems unlikely that this could remedy all concerns and fulfil the conditions of Article 101(3).

Example 2 Situation: A small research company A which does not have its own marketing organisation has discovered and patented a pharmaceutical substance based on new technology that will revolutionise the treatment of a certain disease. Company A enters into an R&D agreement with a large pharmaceutical producer B of products that have so far been used for treating the disease. Company B lacks any similar expertise and R&D programme and would not be able to build such expertise within a relevant timeframe. For the existing products company B has a market share of around 75% in all Member States, but patents are expiring over the next five-year period. There exist two other poles of research at approximately the same stage of development using the same basic new technology. Company B will provide considerable funding and know-how for product development, as well as future access to the market. Company B is granted a license for the exclusive production and distribution of the resulting product for the duration of the patent. It is expected that the product could be brought to market in five to seven years. Analysis: The product is likely to belong to a new relevant market. The parties bring complementary resources and skills to the co-operation, and the probability of the product coming to market increases substantially. Although Company B is likely to have considerable market power on the existing market, this power will be decreasing shortly. The agreement will not lead to a loss in R&D on the part of B, as it has no expertise in this area of research, and the existence of other poles of research are likely to eliminate any incentive to reduce R&D efforts. The exploitation rights during the remaining patent period are likely to be necessary for Company B to make the considerable investments needed and Company A has no own marketing resources. The agreement is therefore unlikely to give rise to restrictive effects on competition within the meaning of Article 101(1). Even if there were such effects, it is likely that the conditions of Article 101(3) would be fulfilled.

142. Risk of foreclosure

Example 3 Situation: A small research company A which does not have its own marketing organisation has discovered and patented a new technology that will revolutionise the market for a certain product for which there is a monopoly producer B worldwide as no competitors can compete with B's current technology. There exist two other poles of research at approximately the same stage of development using the same basic new technology. Company B will provide considerable funding and know-how for product development, as well as future access to the market. Company B is granted an exclusive license for the use of the technology for the duration of the patent and commits to funding only the development of Company A's technology.

EN 40 EN

Analysis: The product is likely to belong to a new relevant market. The parties bring complementary resources and skills to the co-operation, and the probability of the product coming to market increases substantially. However, the fact that Company B commits to Company A's new technology may be likely to lead the two competing poles of research to abandon their projects as it could be difficult to receive continued funding once they have lost the most likely potential customer for their technology. In such a situation no potential competitors would be able to challenge B's monopoly position in the future. The foreclosure effect of the agreement would then be likely to be considered to give rise to restrictive effects on competition within the meaning of Article 101(1). In order to benefit from Article 101(3) the parties would have to show that the exclusivity granted would be indispensible to bring the new technology to the market.

143. Impact of R&D co-operation on dynamic product and technology markets

Example 4 Situation: Two engineering companies that produce vehicle components agree to set up a JV to combine their R&D efforts to improve the production and performance of an existing component. They also pool their existing technology licensing businesses in this area, but will continue to manufacture and sell the components separately. The two companies have market shares in Europe of 15% and 20% on the Original Equipment Manufacturer ("OEM") product market. There are two other major competitors together with several in-house research programmes by large vehicle manufacturers. On the world-wide market for the licensing of technology for these products the parties have shares of 20% and 25%, measured in terms of revenue generated, and there are two other major technologies. The product life cycle for the component is typically two to three years. In each of the last five years one of the major companies has introduced a new version or upgrade. Analysis: Since neither company’s R&D effort is aimed at a completely new product, the markets to consider are those for the existing components and for the licensing of relevant technology. The parties’ combined market share on both the OEM market (35%) and, in particular, on the technology market (45%) are quite high. However, the parties will continue to manufacture and sell the components separately. In addition, there are several competing technologies which are regularly improved. Moreover, the vehicle manufacturers, who do not currently licence their technology, are also potential entrants on the technology market and thus constrain the ability of the parties to profitably raise price. To the extent that the JV restricts competition within the meaning of Article 101(1), which appears to be unlikely, it is likely that it would, in any event, fulfil the criteria of Article 101(3).

4. PRODUCTION AGREEMENTS

4.1. Definition and scope

144. Production agreements vary in form and scope. They can either foresee that production is carried out by only one party or by two or more parties. Companies can

EN 41 EN

produce jointly by way of a joint venture, i.e., a jointly controlled company operating one or several production facilities or by looser forms of co-operation in production such as subcontracting agreements where one party (the "contractor") entrusts to another party (the "subcontractor") the production of a good.

145. There are different types of subcontracting agreements. Horizontal subcontracting agreements are concluded between companies operating in the same product market irrespective of whether they are actual or potential competitors. Vertical subcontracting agreements are concluded between companies operating at different levels of the market.

146. Horizontal subcontracting agreements comprise unilateral and reciprocal specialisation agreements as well as subcontracting agreements with a view to expanding production. Unilateral specialisation agreements are agreements between two parties which are active on the same product market(s), by virtue of which one party agrees to fully or partly cease production of certain products or to refrain from producing those products and to purchase them from the other party, who agrees to produce and supply those products. Reciprocal specialisation agreements are agreements between two or more parties which are active on the same products market(s), by virtue of which two or more parties on a reciprocal basis agree to fully or partly cease or refrain from producing certain but different products and to purchase these products from the other parties, who agree to produce and supply them. In case of subcontracting agreements with a view to expanding production the contractor entrusts the subcontractor with the production of a good, while the contractor does not at the same time cease or limit its own production of the good.

147. These guidelines apply to all forms of joint production agreements and horizontal subcontracting agreements. Subject to certain conditions, joint production agreements as well as unilateral and reciprocal specialisation agreements may benefit from the Specialisation Block Exemption Regulation.73

148. Vertical subcontracting agreements are not covered by these guidelines. They fall within the scope of the Guidelines on Vertical Restraints74 and, subject to certain conditions, may benefit from the Block Exemption Regulation on Vertical Restraints.75 In addition, they may be covered by the Subcontracting Notice.76


149. In order to assess the competitive relationship between the co-operating parties, the relevant market(s) directly concerned by the co-operation in production must first be defined, i.e., the markets to which the products subject to the production agreement belong.

150. A production agreement can also have spill-over effects in markets neighbouring the market directly concerned by the co-operation, for instance upstream or downstream

73 Cited in note 4 above. 74 Cited in note 9 above. 75 Cited in note 8 above 76 Commission notice concerning its assessment of certain subcontracting agreements in relation to Article

85(1) of the EEC Treaty, OJ C1, 3.1.1979, p.2.

EN 42 EN

to the agreement (the so-called “spill-over markets”).77 The spill-over markets are likely to be relevant if the markets are interdependent and the parties are in a strong position on the spill-over market.



151. Production agreements, and in particular production joint ventures, may lead the parties to agree directly on output levels and quality, on the price at which the joint venture sells on its products, or on other competitively important parameters. This may restrict competition even if the parties market the products independently and therefore compete in the downstream market.

152. Production agreements may also result in the coordination of the parties’ competitive behaviour as suppliers leading to higher prices, reduced output, innovation, product quality or variety, i.e. a collusive outcome. This can happen if the production agreement increases the parties' commonality of costs to a degree which enables them to reach a collusive outcome, or if the agreement involves an exchange of commercially sensitive information that can lead to a collusive outcome.

153. Production agreements may furthermore lead to anticompetitive foreclosure of third parties in a related market (e.g., the downstream market relying on inputs from the market in which the production agreement takes place). For instance, by gaining enough market power, parties engaging in joint production in an upstream market may be able to raise the price of a key component for a market downstream. Thereby, they could raise the costs of their rivals downstream and, ultimately, force them off the market. This would in turn increase the parties' market power downstream, which could enable them to sustain prices above the competitive level or harm consumers otherwise. Such competition concerns could materialise irrespective of whether the parties to the agreement are competitors on the market in which the co-operation takes place. However, for this kind of foreclosure to have anticompetitive effects, at least one of the parties must have a strong market position in the market where the risks of foreclosure are assessed.


154. Production agreements constitute a restriction of competition by object if they do not truly concern joint production or horizontal subcontracting, but serve as a tool to engage in a disguised cartel, i.e., otherwise prohibited price fixing, output limitation or market allocation.

155. Generally, agreements which involve price-fixing, limiting output or sharing markets or customers have the object of restricting competition within the meaning of Article 101(1). However, in the context of production agreements, this does not apply where:

– the parties agree on the output directly concerned by the production agreement (e.g., the capacity and production volume of a joint venture or the agreed amount of outsourced products); or

77 As also referred to in Article 2(4) of the Merger Regulation, cited in note 1 above.

EN 43 EN

– a production agreement that also provides for the joint distribution of the manufactured products foresees the joint setting of the sales prices for these products, provided that this is necessary for integrating the production and distribution functions of the agreement.

In these two cases an assessment is required as to whether the agreement gives rise to likely restrictive effects on competition within the meaning of Article 101(1). In both scenarios the agreement on output or prices will not be assessed separately, but in light of the overall effects of the production agreement on the market.


156. Whether the possible competition concerns that production agreements can give rise to are likely to materialise in a given case depends on the characteristics of the market in which the agreement takes place, as well as on the nature and market coverage of the co-operation and the product it concerns. These variables determine the likely effects of a production agreement on competition and thereby the applicability of Article 101(1).

157. Whether a production agreement is likely to give rise to restrictive effects on competition depends on the relevant counterfactual, i.e., the situation that would prevail absent the agreement with all its alleged restrictions. Consequently, production agreements between companies which compete on markets on which the co-operation occurs are not likely to have restrictive effects on competition if the co-operation gives rise to a new market, i.e. if the agreement enables the parties to launch of a new product or service, which the parties would not have been able to do otherwise.

158. A production agreement can directly limit competition between the parties to the agreements, lead to a collusive outcome or anticompetitive foreclosure through increasing the companies' market power, their commonality of costs and/or if they involve the exchange of commercially sensitive information. On the other hand, a direct limitation of competition between the parties, a collusive outcome or anticompetitive foreclosure is not likely to occur if the parties to the agreement do not have market power in the market in which the competition concerns are assessed. It is only market power that can enable them to profitably maintain prices above the competitive level, or profitably maintain output, product quality or variety below what would be dictated by competition.

159. In cases where a company with market power in one market co-operates with a potential entrant, e.g., with a supplier of the same product in a neighbouring geographic or product market, the agreement can potentially increase the market power of the incumbent. This can lead to restrictive effects on competition if actual competition in the incumbent's market is already weak and the threat of entry is a major source of competitive constraint.

160. Production agreements which also involve commercialisation functions, such as joint distribution and/or marketing, carry a higher risk of restrictive effects on competition than mere joint production agreements. Joint commercialisation brings the co-operation closer to the consumer and usually involves the joint setting of prices and sales, i.e., practices that carry the highest risks for competition.

EN 44 EN

161. Nevertheless, commercialisation agreements that are part of a joint production agreement are generally less likely to restrict competition than stand alone commercialisation agreements. For example, in case of an integrated joint production and joint commercialisation agreement the setting of sales prices for the products subject to the agreement does not necessarily restrict competition, provided that the price-fixing by the joint venture is necessary for integrating the production and commercialisation functions of the co-operation agreement. Similarly, where the parties agree on the output directly concerned by a production agreement (e.g. the capacity and production volume of a joint venture or the agreed amount of outsourced products), the agreement is not necessarily restrictive of competition and will have to be analysed in light of its overall effects on the market in order to determine the applicability of Article 101(1).

Market power

162. A production agreement is unlikely to lead to restrictive effects on competition if the parties to the agreement do not have market power in the market on which a restriction of competition is assessed. The starting point for the analysis of market power is the market share of the parties. This will normally be followed by the concentration ratio and the number of players as well as by other dynamic factors such as potential entry, and changing market shares.

163. Companies are unlikely to have market power below a certain level of market share. Therefore, unilateral or reciprocal specialisation agreements as well as joint production agreements including certain integrated commercialisation functions such as joint distribution are block exempted if they are concluded between parties with a combined market share not exceeding 20% in the relevant market(s), provided that the other conditions for the application of the Specialisation Block Exemption Regulation78 are fulfilled. However, if the parties' combined market share exceeds 20%, the restrictive effects have to be analysed as the agreement does not fall within the scope of the Block Exemption Regulation.

164. A moderately higher market share than allowed for in the Specialisation Block Exemption Regulation does not necessarily imply a highly concentrated market, which is an important factor in the assessment. A combined market share of the parties of slightly more than 20% may occur in a market with a moderate concentration. Generally, a production agreement in a concentrated market is more likely to lead to restrictive effects on competition than in a market which is not concentrated. Similarly, a production agreement in a concentrated market may increase the risk of a collusive outcome even if the parties only have a moderate combined market share.

165. Even if the market shares of the parties to the agreement and the market concentration are high, the risks of restrictive effects on competition may still be low if the market is dynamic, i.e., a market with entry and market positions changing frequently.

78 Cited in note 4 above.

EN 45 EN

166. In the analysis of whether the parties to a production agreement have market power, the number and intensity of links (e.g., other co-operation agreements) between the competitors in the market are relevant to the assessment.

167. Factors such as whether the parties to the agreement have high market shares, whether they are close competitors, whether the customers have limited possibilities of switching suppliers, whether competitors are unlikely to increase supply if prices increase, and whether one of the parties to the agreement is an important competitive force, are all relevant for the competitive assessment of the agreement.

Direct limitation of competition between the parties

168. Competition between the parties to a production agreement can be directly limited in various ways. The parties to a production joint venture could, for instance, limit the output of the joint venture compared to what the parties would have brought to the market if each of them had decided their output on their own. Another example would be a joint venture charging a high transfer price to the parties, thereby increasing the input costs for the parties which could lead to higher downstream prices. Competitors may find it profitable to increase their prices as a response, thereby contributing to price increases in the relevant market.

Commonality of costs

169. A production agreement between parties with market power is likely to have restrictive effects on competition if it increases their commonality of costs (i.e., the proportion of costs in common) to a level which enables them to collude. The relevant costs are the variable costs of the product with which the parties to the production agreement compete.

170. A production agreement is more likely to lead to a collusive outcome if prior to the agreement the parties already have a high proportion of variable costs in common as the additional increment (i.e., the production costs of the product subject to the agreement) can tip the balance towards a collusive outcome. Conversely, if the increment is large, the risk of a collusive outcome may be high even if the initial level of commonality of costs is low.

171. Commonality of costs increases the risk of a collusive outcome only if production costs constitute a large proportion of the variable costs concerned. This is, for instance, not the case where the co-operation concerns products which require costly commercialisation. An example would be new or heterogeneous products requiring expensive marketing or high transport costs.

172. Another scenario where commonality of costs can lead to a collusive outcome could be where the parties agree on the joint production of an intermediate product which accounts for a large proportion of the variable costs of the final product with which the parties compete downstream. The parties could use the production agreement to increase the price of this common important input for their products in the downstream market. This would weaken competition downstream and would be likely to lead to higher final prices. The profit would be shifted from downstream to upstream to be then shared cooperatively through the joint venture.

EN 46 EN

173. Similarly, commonality of costs increases the anticompetitive risks of a horizontal subcontracting agreement where the input which the contractor purchases from the subcontractor accounts for a large proportion of the variable costs of the final product with which the parties compete.

Information exchange

174. Any negative effects arising from the exchange of information will not be assessed separately but in the light of the overall effects of the agreement. A production agreement can give rise to restrictive effects on competition if it involves an exchange of commercially strategic information that can lead to a collusive outcome or anticompetitive foreclosure. Whether the exchange of information in the context of a production agreement is likely to lead to restrictive effects on competition should be assessed according to the guidance given in the chapter on the general principles for the competitive assessment of information exchange.

175. If the information exchange does not exceed the sharing of data necessary for the joint production of the goods subject to the production agreement, then even if the information exchange had restrictive effects on competition within the meaning of Art 101(1), the agreement would be more likely to be meet the criteria of Art 101(3) than if the exchange went beyond what was necessary for the joint production. In this case the efficiency gains stemming from producing jointly are likely to outweigh the restrictive effects of the coordination of the parties' conduct. Conversely, in the context of a production agreement the sharing of data which is not necessary for producing jointly, for example the exchange of information related to prices and sales, is less likely to fulfil the conditions of Article 101(3).



176. Production agreements can be pro-competitive if they provide efficiency gains in the form of cost savings or better production technologies. By producing together companies can save costs that otherwise they would duplicate. They can also produce at lower costs if the co-operation enables them to increase production where marginal costs decline with output, i.e., by economies of scale. Producing jointly can also help companies to improve product quality if they put together their complementary skills and know how. Co-operation can also enable companies to increase product variety which they could not afford or would not have been able to achieve otherwise. If joint production allows the parties to increase the number of different types of products, it can also provide cost savings by means of economies of scope.


177. Restrictions that go beyond what is necessary to achieve the efficiency gains generated by a production agreement do not fulfil the criteria of Article 101(3). For instance, restrictions imposed in a production agreement on the parties' competitive conduct with regard to output outside the co-operation will normally not be considered to be indispensible. Similarly, setting prices jointly will not be considered indispensible if the production agreement does not also involve joint commercialisation.

EN 47 EN


178. Efficiency gains attained by indispensible restrictions need to be passed on to consumers in the form of lower prices or better product quality or variety to an extent that outweighs the restrictive effects on competition. Efficiency gains that only benefit the parties or cost savings that are caused by output reduction or market allocation are no sufficient basis for meeting the criteria of Article 101(3). If the parties to the production agreement achieve savings in their variable costs they are more likely to pass them on to consumers than if they reduce their fixed costs. Moreover, the higher the market power of the parties, the less likely they will pass on the efficiency gains to consumers to an extent that would outweigh the restrictive effects on competition.


179. The criteria of Article 101(3) cannot be met if the parties are afforded the possibility of eliminating competition in respect of a substantial part of the products in question. This has to be analysed in the relevant market to which the products subject to the co-operation belong and in possible spill-over markets.

4.5. Examples

180. Creating a new market- counterfactual

Example 1 Situation: Two companies A and B jointly produce and market a product that none of them could have produced and marketed individually. Analysis: Even though they do set quantity and/or price of the new product, there is no restriction of competition because the market would not exist without the co-operation.

181. Commonality of costs and collusive outcomes

Example 2 Situation: Two suppliers A and B of a product X decide to close their current and aged production plants and build a larger, modern and more efficient production plant run by a joint venture, which will have a higher capacity than the total capacity of the old plants of A and B. No other such investments are planned by competitors, who are using their facilities at full capacity. A and B have market shares of 20% and 25% respectively. Their products are the closest substitutes in a specific segment of the market, which is concentrated. The market is rather stagnant, there is no entry and the market shares have been stable over time. Production costs constitute the major part of A and B's variable costs of X, i.e., commercialisation costs of X are minor (marketing costs are low as the good is homogenous and established, transport costs are not significant). Analysis: It is likely that the production joint venture of A and B would give rise to restrictive effects on competition within the meaning of Article 101(1) on the market of X due to the pre-existing market power of A and B, the close substitutability of their products, and the commonality of costs. However, the

EN 48 EN

replacement of two smaller old production plants by the larger, modern and more efficient one may lead the joint venture to increase output at lower prices to the benefits of consumers. However, the production agreement could only meet the criteria of Article 101(3) if the parties provided substantiated evidence that the efficiency gains would be passed on to consumers to such an extent that they would outweigh the restrictive effects on competition.

182. Links between competitors and collusive outcomes

Example 3 Situation: Two suppliers, A and B, form a production joint venture with respect to product Y. A and B each have a 15% market share on the market for Y. There are 3 other players on the market: C with a market share of 30%, D with 25% and E with 15%. B already has a joint production plant with D. Analysis: The market is characterised by very few players and rather symmetric structures. Co-operation between A and B would add an additional link in the market, de facto increasing the concentration in the market, as it would also link D to A and B. This co-operation is likely to increase the risk of a collusive outcome and thereby likely to give rise to restrictive effects on competition within the meaning of Article 101(1). The criteria of Article 101(3) could only be fulfilled in the presence of significant efficiency gains.

183. Anticompetitive foreclosure on a downstream market

Example 4 Situation: A and B set up a production joint venture for the intermediate product X which covers their entire production of X. The production costs of X account for 70% of the variable costs of the final product Y with which A and B compete downstream. A and B each have a share of 20% on the market for Y, there is limited entry and the market shares have been stable over time. In addition to covering their own demand for X, both A and B each have a market share of 40% on the merchant market for X. There are high barriers to entry on the market for X and existing producers are operating near full capacity. On the market for Y, there are two other significant suppliers, each with a 15% market share, and several smaller competitors. This agreement generates economies of scale. Analysis: By virtue of the production joint venture, A and B would be able to largely control supplies of the essential input X to their competitors in the market for Y. This would give A and B the ability to raise their rivals' costs by artificially increasing the price of X, or by reducing the output. This could foreclose the competitors of A and B in market for Y. Because of the likely anticompetitive foreclosure downstream, this agreement is likely to give rise to restrictive effects on competition within the meaning of Article 101(1). The economies of scale generated by the production joint venture are unlikely to outweigh the restrictive effects on competition and therefore this agreement would most likely not meet the criteria of Article 101(3).

184. Specialisation agreement as market allocation

Example 5

EN 49 EN

Situation: A and B each manufacture both products X and Y. A’s market share of X is 30% and of Y 10%. B’s market share of X is 10% and of Y 30%. To obtain economies of scale they conclude a reciprocal specialisation agreement under which A will only produce X and B only Y. They do not cross-supply the good to each other so that A only sells X and B sells only Y. The parties claim that by specialising in this way they save costs due to the economies of scale and by focusing on only one product will improve their production technologies, which will lead to better quality products. Analysis: With regard to its effects on competition in the market, this specialisation agreement is close to a hardcore cartel where parties allocate the market among each other. Therefore, this agreement restricts competition by object within the meaning of Article 101(1). Because the claimed efficiencies in the form of economies of scale and improving production technology are only linked to the market allocation, they are unlikely to outweigh the restrictive effects, and therefore the agreement would not meet the criteria of Article 101(3). In any event, if A or B believes that it would be more efficient to focus on only one product, it can simply take the unilateral decision to only produce X or Y without at the same time agreeing that the other company will focus on producing the respective other product. The analysis would be different if A and B supplied each other with the product they focus on so that they both continue to sell X and Y. In such a case A and B could still compete on price on both markets, especially if production costs (which become common through the production agreement) did not constitute a major share of the variable costs of their products. The relevant costs in this context are the commercialisation costs. Hence, the specialisation agreement would be unlikely to restrict competition if X and Y were largely heterogeneous products with a very high proportion of marketing and distribution costs (e.g., 65-70% or more of total costs). In such a scenario the risks of a collusive outcome would not be high and the criteria of Article 101(3) may be fulfilled, provided that the efficiency gains would be passed on to consumers to such an extent that they would outweigh the restrictive effects on competition of the agreement.

185. Potential competitors

Example 6 Situation: A produces the final product X and B produces the final product Y. X and Y constitute two separate product markets, in which A and B respectively have strong market power. Both companies use Z as an input for their production of X and Y and they both produce Z for captive use only. X is a low added value product for which Z is an essential input (X is quite a simple transformation of Z). Y is a high value added product, for which Z is one of many inputs (Z constitutes a small part of variable costs of Y). A and B agree to jointly produce Z, which generates modest economies of scale. Analysis: A and B are not actual competitors with regard to X, Y or Z. However, since X is a simple transformation of input Z, it is likely that B could easily enter the market of X and thus challenge A's position on that market. The joint production agreement with regard to Z might reduce B's incentives to do so as the joined production might be used for side payments and limit the probability of B to sell product X (as A is likely to have control over the quantity of Z purchased by B from

EN 50 EN

the joint venture). However, the probability of entry of B in the market for X absent the agreement depends on the expected profitability of the entry. As X is a low added value product, entry might not be profitable and thus entry by B could be unlikely in the absence of the agreement. Given that A and B already have market power, the agreement is likely to give rise to restrictive effects on competition within the meaning of Article 101(1) if the agreement does indeed decrease the likelihood of entry of B into A's market, i.e., the market for X. The efficiency gains in the form of economies of scale generated by the agreement are modest and therefore unlikely to outweigh the restrictive effects on competition.

186. Information exchange in a production agreement

Example 7 Situation: Companies A and B with high market power decide to produce together to become more efficient. In the context of this agreement they secretly exchange information about their future prices. The agreement does not cover joint distribution. Analysis: This information exchange makes a collusive outcome likely and is therefore likely to restrict competition by object within the meaning of Article 101(1). It would be unlikely to meet the criteria of Article 101(3) because the sharing of information about the parties' future prices is not indispensable for producing jointly and attaining the corresponding cost savings.

187. Swaps and information exchange

Example 8 Situation: A and B both produce Z, a commodity chemical. Z is a homogenous product which is manufactured according to a European standard which does not allow for any product variations. Production costs are a significant cost factor regarding Z. A has a market share of 20% and B of 25% on the EU-wide market for Z. There are four other manufacturers on the market for Z, with respective market shares of 20%, 15%, 10% and 10%. The production plant of A is located in Member State X in northern Europe whereas the production plant of B is located in Member State Y in southern Europe. Even though the majority of A's customers are located in northern Europe, A also has a number of customers in southern Europe. The same is true for B, which has a number of customers located in northern Europe. Currently, A provides its southern European customers with Z manufactured in its production plant in X and transports it to southern Europe by truck. Respectively, B provides its northern European customers with Z manufactured in Y and transports it to northern Europe also by truck. Transport costs are quite high, but not so high as to make the deliveries by A to southern Europe and B to northern Europe unprofitable. Transport costs from X to southern Europe are lower than from Y to northern Europe. A and B decide that it would be more efficient if A stopped transporting Z from X to southern Europe and if B stopped transporting the Z from Y to northern Europe whereas at the same time they are keen on retaining their customers. To do so, A and B intend to enter into a swap agreement which allows them to purchase an agreed annual quantity of Z from the other party's plant with a view to selling the purchased Z to those of their customers which are located closer to the other party's

EN 51 EN

plant. In order to calculate a purchase price which does not favour one party over the other and which takes due account of the parties' different production costs and different savings on transport costs, and in order to ensure that both parties can achieve an appropriate margin, they agree to disclose to each other their main costs with regard to Z (i.e., production costs and transport costs). Analysis: The fact that A and B – who are competitors – swap parts of their production does not in itself give rise to competition concerns. However, the envisaged swap agreement between A and B foresees the exchange of both parties' production and transport costs with regard to Z. Moreover, A and B have a strong combined market position in a fairly concentrated market for a homogenous commodity product. Therefore, due to the extensive information exchange on a key parameter of competition with regard to Z, it is likely that the swap agreement between A and B will give rise to restrictive effects on competition within the meaning of Article 101(1) as it can lead to a collusive outcome. Even though the agreement will give rise to significant efficiency gains in the form of cost savings for the parties, the restrictions on competition generated by the agreement are not indispensible for their attainment. The parties could achieve similar cost savings by agreeing on a price formula which does not entail the disclosure of their production and transport costs. Consequently, in its current form the swap agreement does not fulfil the criteria of Article 101(3).

188. Market power with low market shares – withdrawal of Specialisation Block Exemption Regulation

Example 9 Situation: A and B agree to enter into a production joint venture with regard to the commodity product X, which does not entail any joint commercialisation. To this end, A and B will close down their existing production plants and build a new production plant with a higher capacity than their existing plants, thereby reducing their production costs. However, A and B could each build a new, modern and efficient production plant by themselves. A and B each have a market share of 9%. There are a large number of smaller producers of X, each with market shares between 1 and 5%. All these producers are operating their plants at full capacity and are unlikely to expand their plants or build new plants in the foreseeable future despite rising demand. Due to the production joint venture, A and B will have around 80% of their relevant costs in common. Customers are fragmented and cannot exert any buyer power. Analysis: Despite a combined market share of merely 18%, the planned production joint venture between A and B is likely to give rise to restrictive effects on competition within the meaning of Article 101(1). Due to the significant commonality of cost caused by the production joint venture it is likely to facilitate a collusive outcome. As all the other producers are rather small and capacity constrained, they could not defeat a price increase by A and B. The same is true for customers as they do not have any countervailing buyer power. However, the production joint venture fulfils the criteria of the Specialisation Block Exemption Regulation, in particular, as the parties' combined market share does not exceed 20%. Nevertheless the Commission would withdraw the benefit of the Specialisation Block Exemption Regulation in this case as the production joint venture is unlikely to fulfil the criteria of Article 101(3). Even though the

EN 52 EN

production joint venture gives rise to cost savings by the parties, in light of their combined market power it is unlikely that they would pass on these efficiency gains to consumers to an extent that would outweigh the restrictive effects on competition.

5. PURCHASING AGREEMENTS

5.1. Definition

189. This chapter focuses on agreements concerning the joint purchase of products. Joint purchasing can be carried out by a jointly controlled company, by a company in which many other companies hold non-controlling stakes, by a contractual arrangement or by even looser forms of co-operation ("joint purchasing arrangements"). Joint purchasing arrangements usually aim at the creation of buying power which can lead to lower prices for consumers. However, buying power may, under certain circumstances, also give rise to competition concerns.

190. Joint purchasing arrangements may involve both horizontal and vertical agreements. In these cases a two-step analysis is necessary. First, the horizontal agreements between the companies engaging in joint purchasing have to be assessed according to the principles described in these guidelines. If this assessment leads to the conclusion that the joint purchasing agreement does not give rise to competition concerns, a further assessment will be necessary to examine the vertical agreements subsequently concluded. The latter assessment will follow the rules of the Block Exemption Regulation and the Guidelines on Vertical Restraints79.

191. A common form of joint purchasing arrangement is an "alliance", that is to say an association of undertakings formed by a group of retailers for the joint purchasing of products. Horizontal agreements concluded between the members of the alliance or decisions adopted by the alliance first have to be assessed as a horizontal co-operation agreement according to these guidelines. Only if this assessment does not reveal any competition concerns it becomes relevant to assess the subsequent vertical agreements between the alliance and an individual member thereof and between the alliance and suppliers. These agreements are covered – subject to certain conditions – by the Block Exemption Regulation on Vertical Restraints. Vertical agreements not covered by this block exemption regulation are not presumed to be illegal but require individual examination.


192. There are two markets which may be affected by joint purchasing arrangements. First, the market(s) with which the joint purchasing arrangement is directly concerned, i.e. the relevant purchasing market(s). Secondly, the selling market(s), i.e., the market(s) downstream where the parties to the joint purchasing arrangement are active as sellers.

193. The definition of relevant purchasing markets follows the principles described in the Market Definition Notice80 and is based on the concept of substitutability to identify competitive constraints. The only difference to the definition of “selling markets” is

79 Cited in notes 8 and 9 above. 80 Cited in note 5 above.

EN 53 EN

that substitutability has to be defined from the viewpoint of supply and not from the viewpoint of demand. In other words: the suppliers’ alternatives are decisive in identifying the competitive constraints on purchasers. These could be analysed, for instance, by examining the suppliers’ reaction to a small but non-transitory price decrease. Once the market is defined, the market share can be calculated as the percentage of the purchases by the parties out of the total sales of the purchased product(s) in the relevant market.

194. If the parties are in addition competitors on one or more selling markets, these markets are also relevant for the assessment. The selling markets have to be defined by applying the methodology described in the Market Definition Notice.



195. Joint purchasing arrangements may lead to restrictive effects on competition on the purchasing and/or downstream selling market(s) such as increased prices, reduced output, product quality or variety, market allocation, or anticompetitive foreclosure of other possible purchasers.

196. If downstream competitors purchase a significant part of their products together, their incentives for price competition on the selling market(s) may be considerably reduced. In case the parties have a significant degree of market power (which does not necessarily amount to dominance) on the selling market(s), the lower purchase prices achieved by the joint purchasing arrangement are likely not to be passed on to consumers.

197. If the parties have a significant degree of market power on the purchasing market (buying power) there is a risk that they may force suppliers to reduce the range or quality of products they produce, which may bring about restrictive effects on competition such as quality reductions, lessening of innovation efforts, or ultimately sub-optimal supply.

198. Buying power of the parties to the joint purchasing arrangement could be used to foreclose competing purchasers by limiting their access to efficient suppliers. This is only possible if there are a limited number of suppliers and there are barriers to entry on the supply side of the upstream market.

199. In general, however, joint purchasing arrangements are less likely to give rise to competition concerns when the parties do not have market power on the selling market(s).


200. Purchasing agreements constitute a restriction of competition by object if they do not truly concern joint purchasing, but serve as a tool to engage in a disguised cartel, i.e., otherwise prohibited price fixing, output limitation or market allocation.

EN 54 EN

201. Agreements which involve the fixing of purchase prices can have the object of restricting competition within the meaning of Article 101(1).81 However, in the context of purchasing agreements this does not apply where the parties to a joint purchasing arrangement agree on the purchasing prices the joint purchasing arrangement may pay to its suppliers for the products subject to the supply contract. In this case an assessment is required as to whether the agreement gives rise to likely restrictive effects on competition within the meaning of Article 101(1). In both scenarios the agreement on purchase prices will not be assessed separately, but in light of the overall effects of the purchasing agreement on the market.


202. Purchasing agreements which do not restrict competition by object must be analysed in their legal and economic context with regard to their actual and likely effects on competition. The analysis of the restrictive effects on competition generated by a purchasing agreement must cover both the negative effects on the purchasing and the selling markets.

Market power

203. There is no absolute threshold above which it can be presumed that a purchasing agreement creates market power and is likely to give rise to restrictive effects on competition within the meaning of Article 101(1). However, in most cases it is unlikely that market power exists if the parties to the purchasing agreement have a combined market share not exceeding 15% on the purchasing market(s) as well as a combined market share not exceeding 15% on the selling market(s). In any event, if the parties' combined market shares do not exceed 15% on both the purchasing as well as the selling market(s), it is likely that the purchasing agreement fulfils the conditions of Article 101(3).

204. A market share above one or both of these thresholds does not automatically indicate that the purchasing agreement is likely to give rise to restrictive effects on competition. A purchasing agreement which does not fall within this safe harbour requires a detailed assessment of its effects on the market involving, but not limited to, factors such as market concentration and possible countervailing power of strong suppliers.

205. Buying power may, under certain circumstances, cause restrictive effects on competition. Anticompetitive buying power is likely to arise if a purchasing agreement accounts for a sufficiently large proportion of the total volume of a purchasing market so that access to the market may be foreclosed to competing purchasers. A high degree of buying power may indirectly affect the output, quality and variety of products on the selling market.

206. In the analysis of whether the parties to a joint purchasing arrangement have buying power, the number and intensity of links (e.g., other purchasing agreements) between the competitors in the market are relevant to the assessment.

81 See Article 101(1) lit. a; Joined Cases T-217/03 and T-245/03, French Beef, cited in note 21 above,

paragraphs 83 et seqq.; Case C-8/08, T-Mobile Netherlands, cited in note 41 above, paragraph 37.

EN 55 EN

207. If, however, competing purchasers co-operate who are not active on the same relevant selling market (e.g., retailers which are active in different geographic markets and cannot be regarded as potential competitors), the agreement is unlikely to have restrictive effects on competition unless the parties have a position in the purchasing markets that could likely be used to harm the competitive position of other players in their respective selling markets.


208. Purchasing agreements may lead to a collusive outcome if they facilitate the coordination of the parties' behaviour on the selling market. This can be the case if the parties achieve a high degree of commonality of costs through joint purchasing.

209. Restrictive effects on competition are more likely if the parties to the purchasing agreement have a significant proportion of their variable costs in the relevant downstream market in common. This is, for instance, the case if retailers which are active in the same relevant retail market(s) jointly purchase a significant amount of the products they offer for resale. It may also be the case if competing manufacturers and sellers of a final product jointly purchase a high proportion of their input together.


210. The implementation of a purchasing agreement may require the exchange of commercially sensitive information such as purchase prices and volumes. The exchange of such information may facilitate coordination with regard to sales prices and output and thus lead to a collusive outcome on the selling markets. Spill-over effects from the exchange of commercially sensitive information can be minimised where data is collated by a joint purchasing organisation which does not pass on the information to its members.

211. Any negative effects arising from the exchange of information will not be assessed separately but in the light of the overall effects of the agreement. Whether the exchange of information in the context of a purchasing agreement is likely to lead to restrictive effects on competition should be assessed according to the guidance given in the chapter on the general principles for the competitive assessment of information exchange. If the information exchange does not exceed the sharing of data necessary for the joint purchasing of the goods subject to the purchasing agreement, then even if the information exchange had restrictive effects on competition within the meaning of Article 101(1), the agreement would be more likely to meet the criteria of Article 101(3) than if the exchange went beyond what was necessary for the joint purchasing.



212. Purchasing agreements can give rise to significant efficiency gains. In particular, purchasing agreements can lead to cost savings such as lower purchase prices or reduced transaction, transportation and storage costs, thereby facilitating economies of scale.

EN 56 EN


213. Restrictions that go beyond what is necessary to achieve the efficiency gains generated by a purchasing agreement do not meet the criteria of Article 101(3). An obligation to purchase exclusively through the co-operation may, in certain cases, be indispensable to achieve the necessary volume for the realisation of economies of scale. However, such an obligation has to be assessed in the context of the individual case.


214. Efficiency gains attained by indispensible restrictions must be passed on to consumers to an extent that outweighs the restrictive effects of competition caused by the purchasing agreement. Hence, cost savings that only benefit the parties to the purchasing agreement will not suffice. Cost savings need to be passed on to consumers, i.e., the parties' customers, e.g., in the form of lower prices on the selling markets. Lower purchasing prices resulting from the mere exercise of buying power are not likely to be passed on to consumers if the purchasers together have market power on the selling markets, and thus do not meet the criteria of Article 101(3). Moreover, the higher the market power of the parties on the selling market(s) the less likely they will pass on the efficiency gains to consumers to an extent that would outweigh the restrictive effects on competition.


215. The criteria of Article 101(3) cannot be fulfilled if the parties are afforded the possibility of eliminating competition in respect of a substantial part of the products in question. This assessment has to cover purchasing and selling markets.

5.5. Examples

216. Joint purchasing by small companies with moderate combined market shares

Example 1 Situation: 150 small retailers conclude an agreement to form a joint purchasing organisation. They are obliged to purchase a minimum volume through the organisation, which accounts for roughly 50% of each retailer’s total costs. The retailers can purchase more than the minimum volume through the organisation, and they may also purchase outside the co-operation. They have a combined market share of 20% on both the purchasing and the selling markets. A and B are their two large competitors. A has a 25% share on both the purchasing and selling markets, B 35%. There are no barriers which would prevent the remaining smaller competitors from also forming a purchasing group. The 150 retailers achieve substantial cost savings by virtue of purchasing jointly through the purchasing organisation. Analysis: The retailers have a moderate market position on the purchasing and the selling markets. Furthermore, the co-operation brings about some economies of scale. Even though the retailers achieve a high degree of commonality of costs, they are unlikely to have market power on the selling market due to the market presence of A and B, which are both individually larger than the joint purchasing organisation. Consequently, the retailers are unlikely to coordinate their behaviour

EN 57 EN

and to reach a collusive outcome. The formation of the joint purchasing organisation is therefore unlikely to give rise to restrictive effects on competition within the meaning of Article 101(1).

217. Commonality of costs and market power on selling market

Example 2 Situation: Two supermarket chains conclude an agreement to jointly purchase products which account for roughly 80% of their variable costs. On the relevant purchasing markets for the different categories of products the parties have combined market shares between 25% and 40%. On the relevant selling market they have a combined market share of 60%. There are four other significant retailers each with a 10% market share. Market entry is not likely. Analysis: It is likely that this purchasing agreement would give the parties the ability to coordinate their behaviour on the selling market, thereby leading to a collusive outcome. The parties have market power on the selling market and the purchasing agreement gives rise to a significant commonality of costs. Moreover, market entry is unlikely. The incentive for the parties to coordinate their behaviour would be reinforced if their cost structures were already similar prior to concluding the agreement. Moreover, similar margins of the parties would further increase the risk of a collusive outcome. This agreement also creates the risk that by the parties' withholding demand and, consequently, as a result of reduced quantity, downstream selling prices would increase. Hence, the purchasing agreement is likely to give rise to restrictive effects on competition within the meaning of Article 101(1). Even though the agreement is very likely to give rise to efficiency gains in the form of cost savings, due to the parties' significant market power on the selling market, these are unlikely to be passed on to consumers to an extent that would outweigh the restrictive effects on competition. Therefore, the purchasing agreement is unlikely to fulfil the criteria of Article 101(3).

218. Parties active in different geographic markets

Example 3 Situation: Six large retailers, which are each based in a different Member State, form a buying alliance to purchase several branded durum wheat flour-based products jointly. The parties are allowed to purchase other similar branded products outside the co-operation. Moreover, five of them also offer similar private label products. The members of the purchasing alliance have a combined market share of approximately 20% on the relevant purchasing market, which is European-wide. In the purchasing market there are three other large players of similar size. Each of the parties to the purchasing alliance has a market share between 20% and 40% on the national selling markets on which they are active. None of them is active in a Member State where another member of the alliance is active. The parties are not potential entrants to each others' markets. Analysis: The purchasing alliance will be able to compete with the other existing major players on the purchasing market. The selling markets are much smaller (in turnover and geographic scope) than the European-wide purchasing market and in these markets some of the members of the alliance may have market power. Even if the members of the purchasing alliance have a combined market share of more than

EN 58 EN

15% on the purchasing market, the parties are unlikely to coordinate their conduct and collude on the selling markets since they are neither actual nor potential competitors on the downstream markets.

219. Information exchange

Example 4 Situation: Three competing manufacturers A, B and C entrust an independent joint purchasing organisation with the purchase of product Z, which is an intermediary product used by the three parties for their production of the final product X. The costs of Z are not a significant cost factor for the production of X. The joint purchasing organisation does not compete with the parties on the selling market for X All information necessary for the purchases (e.g. quality specifications, quantities, delivery dates, maximum purchase prices) is only disclosed to the joint purchasing organisations, not to the other parties. The joint purchasing organisation agrees the purchasing prices with the suppliers. A, B and C have a combined market share of 30 % on each of the purchasing and selling markets. They have six competitors in the purchasing and selling markets, two of them have a market share of 20%. Analysis: Since there is no direct information exchange between the parties, the transfer of the information necessary for the purchases to the joint purchasing organisation is unlikely to lead to a collusive outcome. Consequently, the exchange of information is unlikely to give rise to restrictive effects on competition within the meaning of Article 101(1).

6. AGREEMENTS ON COMMERCIALISATION

6.1. Definition

220. Commercialisation agreements involve co-operation between competitors in the selling, distribution or promotion of their substitute products. This type of agreements can have a widely varying scope, depending on the marketing functions which are being covered by the co-operation. At one end of the spectrum, there is joint selling that leads to a joint determination of all commercial aspects related to the sale of the product, including price. At the other end, there are more limited agreements that only address one specific marketing function, such as distribution, after-sales service, or advertising.

221. An important category of these more limited agreements is distribution agreements. The Block Exemption Regulation on Vertical Restraints82 and Guidelines on Vertical Restraints83 generally cover distribution agreements unless the parties to the agreement are actual or potential competitors. If the parties are competitors, the Block Exemption Regulation on Vertical Restraints only covers non-reciprocal vertical agreements between competitors, if (a) the supplier is a manufacturer and a distributor of goods, while the buyer is a distributor and not a competing undertaking at the manufacturing level or, (b) the supplier is a provider of services at several

82 Cited in note 8 above. 83 Cited in note 9 above.

EN 59 EN

levels of trade, while the buyer provides its goods or services at the retail level and does not provide competing services at the level of trade where it purchases the contract services.84

222. If competitors agree to distribute their substitute products on a reciprocal basis (on different geographic markets) there is a possibility in certain cases that the agreements have as their object or effect the partitioning of markets between the parties or that they lead to a collusive outcome. The same is true for non-reciprocal agreements between competitors exceeding a certain size. Reciprocal agreements and non-reciprocal agreements between competitors exceeding a certain size thus have first to be assessed according to the principles set out below. If this assessment leads to the conclusion that a co-operation between competitors in the area of distribution would in principle be acceptable, a further assessment will be necessary to examine the vertical restraints included in such agreements. This second step of the assessment should be based on the principles set out in the Guidelines on Vertical Restraints.85

223. A further distinction should be drawn between agreements where the parties agree only on joint commercialisation and agreements where the commercialisation is related to another type of co-operation upstream, such as joint production or joint purchasing. The analysis of commercialisation agreements combining different stages of co-operation should start on the basis of the chapter of these guidelines dealing with the "most upstream indispensable building block" of the co-operation86.


224. To assess the competitive relationship between the parties, the relevant product and geographic market(s) directly concerned by the co-operation (i.e. the market(s) to which the products subject to the agreement belong) have to be defined. As a commercialisation agreement in one market may also affect the competitive behaviour of the parties in a neighbouring market which is closely related to the market directly concerned by the co-operation, the neighbouring market – if any - also needs to be defined. Such a neighbouring market may be horizontally or vertically related to the market where the co-operation takes place.



225. Commercialisation agreements can lead to restrictions of competition in several ways. First, price fixing is the most obvious restriction that commercialisation agreements may entail, given that in order to jointly market a product, a common price is often an inherent element of the commercialisation, for which there would be a need to undertake a thorough competition assessment.

84 Article 2(4) of Block Exemption Regulation on Vertical Restraints, cited in note 8 above. 85 Cited in note 9 above. 86 See paragraphs 13 and 14 of these guidelines.

EN 60 EN

226. Second, commercialisation agreements may also facilitate output limitation, since by means of the agreement the parties may decide on the volume of products to be put in the market, therefore restricting supply.

227. Third, commercialisation agreements may become a means for the parties to divide the markets or to allocate orders or customers, in particular in those cases where the parties' production plants are located in different geographic markets or when the agreements are reciprocal.

228. Finally, commercialisation agreements could also lead to an exchange of information relating to aspects within or outside the scope of the co-operation or to commonality of costs – in particular with regard to agreements not encompassing price fixing – which may result in a collusive outcome.


229. A major competition concern about a commercialisation agreement between competitors is price fixing. Agreements limited to joint selling generally have the object of coordinating the pricing policy of competing manufacturers or service providers. In this case they not only eliminate price competition between the parties on substitute products but also restrict the total volume of products to be delivered by the parties within the framework of a system for allocating orders. Such agreements are therefore likely to have as their object a restriction of competition within the meaning of Article 101(1).

230. This assessment does not change if the agreement is non-exclusive, i.e., where the parties are free to sell individually outside the agreement, as long as it can be presumed that the agreement will lead to an overall coordination of the prices charged by the parties.

231. Another specific competition concern related to distribution arrangements between parties which are active in different geographic markets is that they can be an instrument of market partitioning. In case the parties use a reciprocal distribution agreement to distribute each other’s products to deliberately allocate markets or customers and eliminate competition between them, the agreement is likely to have as its object a restriction of competition. If the agreement is not reciprocal, the risk of market partitioning is less pronounced. It needs, however, to be assessed whether the non-reciprocal agreement constitutes the basis for a mutual understanding to avoid entering each other's market.


232. A commercialisation agreement is normally not likely to create competition concerns if it is objectively necessary to allow one party to enter a market it could not have entered individually or with a more limited number of parties than the ones effectively taking part in the co-operation, e.g., because of the costs involved. A specific application of this principle would be consortia arrangements that allow the companies involved to participate in projects that they would not be able to undertake individually. As the parties to the consortia arrangement are therefore not potential competitors for implementing the project, there is no restriction of competition within the meaning of Article 101(1).

EN 61 EN

233. Similarly, not all reciprocal distribution agreements have as their object a restriction of competition. Depending on the facts of the case at hand, some reciprocal distribution agreements may, nevertheless, have restrictive effects on competition. The key issue in assessing an agreement of this type is whether the agreement in question is objectively necessary for the parties to enter each other’s market. If it is, the agreement does not create competition problems of a horizontal nature. However, if the agreement reduces the decision-making independence of one of the parties with regard to entering the other parties' market by limiting its incentives to do so, it is likely to give rise to restrictive effects on competition. The same reasoning applies to non-reciprocal agreements, where the risk of restrictive effects on competition is, however, less pronounced.

234. Moreover, the distribution agreement can have restrictive effects on competition if it contains vertical restraints, such as restrictions on passive sales, resale price maintenance, etc.

Market power

235. Commercialisation agreements between competitors can only have restrictive effects on competition if the parties have some degree of market power. In most cases, it is unlikely that market power exists if the parties to the agreement have a combined market share not exceeding 15%. In any event, if the parties' combined market share does not exceed 15% it is likely that the conditions of Article 101(3) explained below are fulfilled by the agreement in question.

236. If the parties’ combined market share is greater than 15%, it falls outside the safe harbour and thus the likely impact of the joint commercialisation agreement on the market must be assessed.


237. A joint commercialisation agreement that does not involve price fixing is also likely to give rise to restrictive effects on competition if it increases the parties' commonality of variable costs to a level which is likely to lead to a collusive outcome. This is likely to be the case for a joint commercialisation agreement if prior to the agreement the parties already have a high proportion of their variable costs in common as the additional increment (i.e., the commercialisation costs of the product subject to the agreement) can tip the balance towards a collusive outcome. Otherwise, if the increment is large, the risk of a collusive outcome may be high even if the initial level of commonality of costs is low.

238. Commonality of costs only increases the risk of a collusive outcome if the parties have market power and if the commercialisation costs constitute a large proportion of the variable costs related to the products concerned. This is, for example, not the case for homogeneous products for which the highest cost factor is production. However, commonality of commercialisation costs increases the risk of a collusive outcome if the commercialisation agreement concerns products which entail costly commercialisation, e.g., high distribution or marketing costs. Consequently, joint advertising or joint promotion agreements can also give rise to restrictive effects on competition in case these costs constitute a significant cost factor.


EN 62 EN

239. Joint commercialisation generally involves the exchange of sensitive commercial information, particularly on marketing strategy and pricing. In most commercialisation agreements, some degree of information exchange is required for implementing the agreement. It is therefore necessary to verify whether the information exchange can give rise to a collusive outcome with regard to the parties' activities within and outside the co-operation. Any negative effects arising from the exchange of information will not be assessed separately but in the light of the overall effects of the agreement.

240. For example, where the parties to a joint advertising agreement exchange pricing information, this may lead to a collusive outcome with regard to the sale of the jointly advertised products. In any event, the exchange of such information in the context of a joint advertising agreement goes beyond what would be necessary to implement this agreement. The likely restrictive effects on competition of information exchange in the context of commercialisation agreements will depend on the characteristics of the market and the data shared, and should be assessed under to guidance given in the chapter on information exchange.



241. Commercialisation agreements can give rise to significant efficiency gains. The efficiencies to be taken into account when assessing whether a commercialisation agreement fulfils the criteria of Article 101(3) will depend on the nature of the activity and the parties to the co-operation. Price fixing can generally not be justified, unless it is indispensable for the integration of other marketing functions, and this integration will generate substantial efficiencies. Joint distribution can generate significant efficiencies, stemming from economies of scale or scope, especially for smaller producers.

242. In addition, the efficiency gains should not be savings which result only from the elimination of costs that are inherently part of competition, but must result from the integration of economic activities. A reduction of transport cost which is only a result of customer allocation without any integration of the logistical system can therefore not be regarded as an efficiency gains within the meaning of Article 101(3).

243. Efficiency gains must be demonstrated. An important element in this respect would be the contribution by the parties of significant capital, technology, or other assets. Cost savings through reduced duplication of resources and facilities can also be accepted. However, if the joint commercialisation represents no more than a sales agency without any investment, it is likely to be a disguised cartel and as such unlikely to fulfil the conditions of Article 101(3).


244. Restrictions that go beyond what is necessary to achieve the efficiency gains generated by a commercialisation agreement do not fulfil the criteria of Article 101(3). The question of indispensability is especially important for those agreements involving price fixing or market allocation, which can only under exceptional circumstances be considered as indispensible.

EN 63 EN


245. Efficiency gains attained by indispensible restrictions must be passed on to consumers to an extent that outweighs the restrictive effects on competition caused by the commercialisation agreement. This can happen in the form of lower prices or better product quality or variety. The higher the market power of the parties, however, the less likely it is that efficiency gains will be passed on to consumers to an extent that outweighs the restrictive effects on competition. In case the parties have a combined market share of below 15%, it is likely that any demonstrated efficiency gains generated by the agreement will be sufficiently passed on to consumers.


246. The criteria of Article 101(3) cannot be fulfilled if the parties are afforded the possibility of eliminating competition in respect of a substantial part of the products in question. This has to be analysed in the relevant market to which the products subject to the co-operation belong and in possible spill-over markets.

6.5. Examples

247. Joint commercialisation necessary to enter a market

Example 1 Situation: Four companies providing laundry services, each with a 3% market share of the overall laundry market in one city, agree to create a joint marketing arm for the selling of laundry service to institutional customers (i.e., hotels, hospitals, offices), whilst keeping their independence and freedom to compete for local, individual clients. In view of the new segment of demand (the institutional customers) they develop a common brand name, a common price and common standard terms including, inter alia, a maximum 24h time-length before deliveries and schedules for delivery. They set up a common call centre where institutional clients can request their collection and/or delivery service. They hire a receptionist (for the call centre) and several drivers. They further invest in vans for dispatching, and in brand promotion, to increase their visibility. The agreement does not fully reduce their individual infrastructure costs (since they are keeping their own premises and still compete with each other for the individual local clients), but it increases their economies of scale and allows them to offer a more comprehensive service to other types of clients, which includes longer opening hours and dispatching to a wider geographic coverage. In order to ensure the viability of the project, it is indispensable that all four of them enter into the agreement. The market is very fragmented, with no individual competitor having more than 15% market share. Analysis: Although the joint market share of the parties is below 15%, the fact that the agreement involves price fixing means that Article 101(1) could apply. However, the parties would not have been in a position to enter the market of providing laundry services to institutional customers, either individually or in co-operation with a fewer number of parties than the four currently taking part in the agreement. As such, the agreement would not create competition concerns, irrespective of the price-fixing restriction, which in this case can be considered as

EN 64 EN

indispensable to the promotion of the common brand and the success of the project.

248. Commercialisation agreement by more parties than necessary to enter a market

Example 2 Situation: The same facts as in Example 1 apply with one main difference: in order to ensure the viability of the project, the agreement could have been implemented by only three of the parties (instead of the four actually taking part in the co-operation). Analysis: Although the joint market share of the parties is below 15%, the fact that the agreement involves price fixing and could have been carried out by fewer than the four parties means that Article 101(1) applies. The agreement thus needs to be assessed under Article 101(3). The agreement gives rise to efficiency gains as the parties are now able to offer improved services for a new category of customers on a larger scale (which they would not otherwise have been able to service individually). In light of the parties' combined market share of below 15%, it is likely that they will sufficiently pass-on any efficiency gains to consumers. It is further necessary to consider whether the restrictions imposed by the agreement are indispensable to achieve the efficiencies and whether the agreement eliminates competition. Given that the aim of the agreement is to provide a more comprehensive service (including the dispatch, which was not offered before) to an additional category of customers, under a single brand with common standard terms, the price fixing can be considered as indispensable to the promotion of the common brand and, consequently, the success of the project and the resulting efficiencies. Additionally, taking into account the market fragmentation, the agreement will not eliminate competition. The fact that there are four parties to the agreement (instead of the three that would have been strictly necessary) allows for increased capacity and contributes to simultaneously fulfil the demand of several institutional customers in compliance with the standard terms (i.e., meeting maximum time delivery terms). As such, the efficiency gains are likely to outweigh the restrictive effects arising from the reduction of competition between the parties and the agreement is likely to fulfil the conditions of Article 101(3).

249. Joint internet platform

Example 3 Situation: A large number of specialty shops throughout a country join an electronic web-based platform for the promotion, sale and delivery of gift fruit baskets. Through a monthly fee, they share the running costs of the platform and jointly invest in brand promotion. Through the webpage, where a number of different types of gift baskets are offered, customers order (and pay) the type of gift basket they want to be delivered. The order is then allocated to the specialty shop closest to the address of delivery. The shop individually bears the costs of composing the gift basket and delivering it to the client. It reaps 90% of the final price, which is set by the web-based platform and uniformly applies to all participating specialty shops, whilst the remaining 10% is devoted to the common promotion and the running costs of the web-based platform. Apart from the payment of the monthly fee, there are no further restrictions for specialty shops to join the platform, throughout the national territory. Moreover, specialty shops having their own company website are also able to sell gift fruit baskets on the internet under

EN 65 EN

their own name and thus can still compete among themselves outside the co-operation through the web-based platform. Analysis: Although the agreement is of a limited nature, since it only covers the joint selling of a particular type of product through a specific marketing channel (the web-based platform) and since it involves price-fixing, it is likely to restrict competition by object within the meaning of Article 101(1). The agreement therefore needs to be assessed under Article 101(3). The agreement gives rise to efficiency gains such as greater choice and higher quality service. These efficiency gains benefit consumers and are likely to outweigh the restrictive effects on competition the agreement brings about. Given that the specialty stores taking part in the co-operation are still able to operate individually and to compete one with another, both through their shops and the internet, the price-fixing restriction could be considered as indispensable for the promotion of the product (since when buying through the web-based platform consumers do not know where they are buying the gift basket from) and the ensuing efficiency gains. In the absence of other restrictions, the agreement fulfils the criteria of Article 101(3).

250. Sales joint venture

Example 4 Situation: Companies A and B, located in two different countries, produce bicycle tyres. They have a combined market share of 14% on the EU-wide market for bicycle tyres. They decide to set up a (non full-function) sales joint venture for marketing the tyres to bicycle producers and agree on selling all their production through the joint venture. The production and transport infrastructure remains separate within each party. The parties claim considerable efficiency gains stemming from the agreement. Such gains would mainly relate to increased economies of scale, being able to fulfil the demands of their existing and potential new customers and better competing with imported tyres produced in third countries. The joint venture negotiates the prices and allocates orders to the closest production plant, as a way to rationalise transport costs when further delivering to the customer. Analysis: Even though the combined market share of the parties is below 15%, the agreement falls under Article 101(1). It restricts competition by object since it involves customer allocation and the setting of prices by the joint venture. The claimed efficiencies deriving from the agreement do not result from the integration of economic activities or from common investment. The joint venture would have a very limited scope and would only serve as an interface for allocating orders to the production plants. It is therefore unlikely that any efficiency gains would be passed on to consumers to such an extent that they would outweigh the restrictive effects on competition brought about by the agreement. As such the conditions of Article 101(3) would not be fulfilled.

251. Non-poaching clause in agreement on outsourcing of services

Example 5 Situation: A and B are both competing providers of cleaning services for commercial premises. Both have a market share of 15%. There are several other competitors with market shares between 10 and 15%. A has taken the (unilateral)

EN 66 EN

decision to only focus on large customers in the future as servicing large and small customers have proved to require a somewhat different organisation of the work. Consequently, A has decided to no longer enter into contracts with new small customers. In addition, A and B enter into an outsourcing agreement whereby B would directly provide cleaning services to A's existing small customers (which represent 1/3 of his customer base). At the same time, A is keen on not losing the customer relationship with these small customers. Hence, A will continue to keep its contractual relationships with the small customers but the direct provision of the cleaning services will be done by B. In order to implement the outsourcing agreement, A will necessarily need to provide B with the identities of A's small customers which are subject to the agreement. As A is afraid that B may try to poach these customers by offering cheaper direct services (thereby by-passing A), A insists that the outsourcing agreement contain a "non-poaching clause". According to this clause, B may not contact the small customers falling under the outsourcing agreements with a view to providing direct services to them. In addition, A and B agree that B may not even provide direct services to these customers if B is approached by them. Without the "non-poaching clause" A would not enter into an outsourcing agreement with B or any other company. Analysis: The outsourcing agreement removes B as an independent supplier of cleaning services for A's small customers as they will no longer be able to enter into a direct contractual relationship with B. However, these customers only represent 1/3 of A's customer base, i.e., 5% of the market. They will still be able to turn to A's and B's competitors, which represent 70% of the market. Hence, the outsourcing agreement will not enable A to profitably raise the prices charged to the customers subject to the outsourcing agreement. In addition, the outsourcing agreement is not likely to give rise to a collusive outcome as A and B only have a combined market share of 30% and they are faced with several competitors that have market shares similar to A's and B's individual market shares. Moreover, the fact that servicing large and small customers is somewhat different minimises the risk of spill-over effects from the outsourcing agreement to A's and B's behaviour when competing for large customers. Consequently, the outsourcing agreement is not likely to give rise to restrictive effects on competition within the meaning of Article 101(1).

7. STANDARDISATION AGREEMENTS

7.1. Definition

Standardisation agreements

252. Standardisation agreements have as their primary objective the definition of technical or quality requirements with which current or future products, production processes, services or methods may comply.87 Standardisation agreements can cover various issues, such as standardisation of different grades or sizes of a particular product or technical specifications in product or services markets where compatibility and interoperability with other products or systems is essential. The terms of access to a

87 Standardisation can take different forms, ranging from the adoption of consensus based standards by the

recognised European or national standards bodies, through consortia and fora, to agreements between independent companies.

EN 67 EN

particular quality mark or for approval by a regulatory body can also be regarded as a standard. Agreements setting out standards on the environmental performance of products or production processes are also covered by this chapter.

253. The preparation and production of technical standards as part of the execution of public powers are not covered by these guidelines.88 The European standards bodies recognised under Directive 98/34/EC89 are subject to competition law to the extent that they can be considered to be an undertaking or an association of undertakings within the meaning of Articles 101 and 102. Standards related to the provision of professional services, such as rules of admission to a liberal profession, are not covered by these guidelines.

Standard terms

254. In certain industries companies use standard terms and conditions of sale or purchase90 elaborated by a trade association or directly by the competing companies ("standard terms"). Such standard terms are covered by these guidelines to the extent that they establish standard conditions of sale or purchase between competitors and consumers (and not the conditions of sale or purchase between competitors) for substitute products. When such standard terms are widely used within an industry, the conditions of purchase or sale used in the industry might, to a large extent, become de facto standardised. Examples of industries in which standard terms play an important role are the banking (e.g., contracts for the issuance of cards, use of ATMs and account terms) and insurance sectors.

255. Standard terms and conditions elaborated individually by a company solely for its own use when contracting with its suppliers or customers are, however, not horizontal agreements and therefore not covered by these guidelines.


256. Standardisation agreements may produce their effects on four possible markets, which will be defined according to the Market Definition Notice.91 First, standard-setting may impact the product or service market(s) to which the standard(s) relate(s). Second, where the standard-setting involves the selection of technology and where the rights to intellectual property are marketed separately from the products to which they relate, the standard can have effects on the relevant technology market.92 Third, the service market for standard-setting may be affected if different standard-setting bodies or agreements exist. Fourth, where relevant, a distinct market for testing and certification may be impacted by standard-setting.

257. As regards standard terms, the effects are, in general, felt on the downstream market where the companies using the standard terms compete by selling their product to their customers.

88 See Case C-113/07, SELEX, [2009] ECR I not yet reported, paragraph 92. 89 Directive 98/34/EC of the European Parliament and of the Council laying down a procedure for the

provision of information in the field of technical standards and regulations, OJ L204, 21.7.1998, p.37. 90 Such standard terms might cover only a very small part of the clauses contained in the final contract or

a large part thereof. 91 Cited in note 5 above. 92 See the chapter on R&D agreements.

EN 68 EN




258. Standardisation agreements generally have a positive economic effect, for example by promoting economic interpenetration on the internal market and encouraging the development of new markets and improved supply conditions. Standards may increase competition and lower output and sales costs, benefiting economies as a whole. Standards may maintain and enhance quality, provide information and ensure interoperability (thus increasing value for consumers).

259. Standard-setting can, however, give rise to restrictive effects on competition by potentially restricting price competition and limiting or controlling production, markets, innovation or technical development. Discussions in the context of standard-setting, like all meetings between competitors, can provide an opportunity to reduce or eliminate price competition in the markets concerned, thereby facilitating a collusive outcome on the market.

260. In addition, standards that set detailed technical specifications for a product or service may limit technical development and innovation. While a standard is being developed, alternative technologies can compete for inclusion in the standard. Once one technology has been chosen and the standard has been set, the competing technologies face a barrier to entry and are potentially excluded from the market.

261. Standard-setting can also give rise to restrictive effects on competition by way of foreclosure or in the form of limitation of innovation when the process for selecting the technologies in the standard is de facto controlled by one or more stakeholders or where the standard-setting process is biased towards one or more participants.

262. As a standard can constitute a barrier to entry, a company (or potentially more than one company) holding essential intellectual property rights (IPR) in a standard could control its use and thereby the product or service market to which the standard relates. This in turn could allow the company in question to abuse its dominant position by extracting excess rents by "holding-up" users after the adoption of the standard. Given the particular risks arising in this context, this chapter focuses on standardisation agreements involving IPR, but the rules apply to all standardisation agreements.93

Standard terms

263. Standard terms can give rise to restrictive effects on competition by limiting product choice and innovation. If a large part of an industry adopts the standard terms and chooses not to deviate from them in individual cases (or only deviates from them in exceptional cases of strong buyer-power), customers might have no option other than to accept the conditions in the standard terms. The risk of limiting choice and innovation is, however, only likely in cases where the standard terms define the

93 Typically standard-setting organisations have IPR policies setting out their procedures relating to IPR.

The rules set out in these IPR policies are of particular importance in the present context.

EN 69 EN

scope of the end-product as, for example in insurance contracts. As regards classical consumer goods, standard terms of sale generally do not limit innovation of the actual product or product quality and variety.

264. In addition, depending on their content, standard terms might risk affecting the commercial conditions of the final product. In particular, there is a serious risk that standard terms relating to price would restrict price competition.

265. Moreover, if the standard terms become industry practice, access to them might be vital for entry into the market. In such cases, refusing access to the standard terms could risk causing anti-competitive foreclosure. As long as the standard terms remain effectively open for use for anyone that wishes to have access to them, they are unlikely to give rise to anti-competitive foreclosure.


266. Agreements that use a standard or standard terms as part of a broader restrictive agreement aimed at excluding actual or potential competitors restrict competition by object within the meaning of Article 101(1). For instance, an agreement whereby a national association of manufacturers sets a standard and puts pressure on third parties not to market products that do not comply with the standard would fall into this category. Another example would be where a trade association does not allow a new entrant access to its standard terms, the use of which is vital to ensure entry to the market.

267. Any efforts to reduce competition by using the disclosure of essential IPR or most restrictive licensing terms prior to the adoption of a standard as a cover to jointly fix prices of products will constitute restrictions of competition by object within the meaning of Article 101(1). Prior to the adoption of the standard, agreements by IPR holders on the licensing terms they will disclose will also constitute restrictions of competition by object within the meaning of Article 101(1).

268. Any standard terms containing provisions which influence the prices charged to customers (i.e., recommended prices, rebates, etc.) would constitute a restriction of competition by object.



269. The assessment of each standardisation agreement must in general take into account the nature of the standard and its likely effects on the markets concerned. Whether standardisation agreements may give rise to restrictive effects on competition depends, among other factors, on the extent to which the members remain free to develop alternative standards or products that do not comply with the agreed standard.

270. In the standard-setting process, some tension is inevitable as each firm desires to promote its own solutions as part of the standard. At the same time, each company also needs to work together with other members of the standard-setting organisation to develop, establish and promote the standard.

EN 70 EN

271. The tensions that may arise in standard-setting processes are magnified by the fact that companies involved in standard-setting might differ in the scope of their economic activities, sources of revenues and thus their incentives.

272. Upstream-only companies solely develop and market technologies. Normally, licensing revenue is their only source of income and hence their incentive is to maximise their royalties.

273. Conversely, downstream-only companies, solely manufacture products based on technologies developed by others and do not hold relevant IPR. As royalties represent a cost for them, and not a source of revenue, their incentive is to reduce royalties.

274. Vertically integrated companies that both develop technology and sell products have mixed incentives. On the one hand, they can draw revenue from their IPR if they so choose. On the other hand, they may have to pay royalties to other companies holding IPR essential to the standard for the products they manufacture. When the bulk of the revenues (and profits) of these companies is generated downstream they are less dependent than upstream-only companies on the revenue they may obtain by licensing their essential IPR. They might therefore have a stronger incentive to cross-license their own essential IPR in exchange or essential IPR held by other companies than in seeking royalties.

275. However, irrespective of the market positioning of the different IPR holders, the establishment of standards, which might take years to complete, can create or increase the market power of those IPR holders and in some circumstances lead to abuses of a dominant position. During the development of the standard, different patented technologies may be in competition with each other for inclusion in the standard. Up until the adoption of a standard, the industry may have flexibility with respect to the exact technical characteristics of the standards, and thus may be able to adjust the standard so that it avoids relying on certain patents. However, once a specific patented technology is included in a standard (and the alternatives rejected), the industry may be locked in, inter alia, because of the costs of reengineering or switching away from the standard.

276. In light of the above, the following paragraphs set out the conditions in which standard-setting agreements fall outside the scope of Article 101(1). It is not necessary for standard-setting agreements to fulfil these conditions, but normally they will be sufficient to avoid the application of Article 101(1). If a standard-setting agreement does not fulfil these conditions, an individual assessment is required to establish whether the agreement falls under Article 101(1) and in the affirmative, whether the conditions of Article 101(3) would be fulfilled such that the agreement falls within the legal exception to the Article 101(1) prohibition.

277. Where participation in standard-setting, as well as the procedure for adopting the standard in question, is unrestricted and transparent, standardisation agreements which set no obligation to comply with the standard and provide access to the standard on fair, reasonable and non-discriminatory terms do not restrict competition within the meaning of Article 101(1).

EN 71 EN

278. First, with respect to unrestricted participation and the procedure for adopting the standard, the rules for the standard-setting organisation, and in particular its IPR policy, should guarantee that all relevant actors can participate in the process leading to the selection of the standard. Notably, the relevant rules should not exclude or discriminate against specific groups of IPR holders. There should be no bias in favour or against royalty free standards, depending on the relative benefits of the latter compared to other alternatives. The standard-setting organisations should also have objective and non-discriminatory procedures for allocating voting rights.

279. Second, with respect to transparency, the relevant standard-setting organisation should have procedures which allow stakeholders to inform themselves of upcoming, on-going and finalised standardisation work.

280. Third, the standard-setting organisation's rules must seek to avoid the misuse of the standardisation process through hold-ups and the charging of abusive royalty rates by IPR holders. These objectives should be ensured in standard-setting organisations through rules which are binding on the standard-setting organisation's members.

281. This requires a clear and balanced IPR policy that protects against companies abusing market power with respect to a standard. Thus, the IPR policy should require good faith disclosure of those intellectual property rights94 that might be essential for the implementation of a standard under development before that standard is agreed. This requires that the IPR holders make reasonable efforts to identify existing and pending IPR reading on the potential standard.

282. The IPR policy should also require that all holders of essential IPR in technology which may be adopted as part of a standard provide an irrevocable commitment in writing to license their IPR to all third parties on fair, reasonable and non-discriminatory terms ("FRAND commitment").

283. The aim of FRAND commitments in the context of standard-setting is to ensure that patented technology incorporated in a standard is accessible to the users of that standard on fair, reasonable and non-discriminatory terms and conditions. In particular, FRAND commitments are intended to prevent IPR holders from making the implementation of a standard difficult by refusing to license or by requesting unfair or unreasonable fees (in other words excessive fees) after the industry has been locked-in to the standard and/or charging discriminatory royalty fees.

284. An abuse of the market power gained by virtue of IPR being included in a standard constitutes an infringement of Article 102. In this context and in case of a dispute, the assessment of whether fees imposed for patents in the standard-setting context are unfair or unreasonable, will be based on whether the fees bear a reasonable relationship to the economic value of the patents.95 Various methods may be available to make this assessment. In principle, cost-based methods are not well adapted to this context because of the difficulty in assessing the costs attributable to the development of a particular patent or groups of patents. Instead , it may be

94 This includes, e.g., patents and patent applications. 95 See Case 27/76, United Brands v Commission, [1978] ECR 207, paragraph 250. See also Case C-

385/07 P, Der Grüne Punkt – Duales System Deutschland GmbH, [2009] ECR I not yet reported, paragraph 142.

EN 72 EN

possible to compare the licensing fees charged by the undertaking in question for the relevant patents in a competitive environment before the industry has been locked into the standard (ex ante) with those charged after the industry has been locked in (ex post). This assumes that the comparison can be made in a consistent and reliable manner.96

285. Another method of assessing the relationship of the IPR fees to the economic value of the patents could be to obtain an independent expert assessment of the relevant IPR portfolio's objective quality and centrality to the standard at issue. It may also be possible to rely on previous unilateral ex ante disclosures of most restrictive licensing terms. This also assumes that the comparison can be made in a consistent and reliable manner. These guidelines do not seek to provide an exhaustive list of appropriate methods to assess whether the royalty fees are excessive.

286. To ensure the effectiveness of the FRAND commitment, there should also be a requirement on all IPR holders who provide such a commitment to take all necessary measures to ensure that any undertaking to which the IPR owner transfers its IPR (including the right to license that IPR) is bound by that commitment.

287. In addition to fulfilling the above conditions that will normally be sufficient to avoid restrictive effects on competition, standard-setting agreements can contain additional rules which may or may not lead to a likely restriction of competition. To take a notable example, if standard-setting organisations provide for ex ante disclosures of most restrictive licensing terms, this will not (subject to the caveat set out above) lead to a restriction of competition within Article 101(1). In this regard, it is important that parties involved in the selection of a standard be fully informed not only as to the available technical options and the associated IPR, but also as to the likely cost of that IPR. Therefore, should a standard-setting organisation's IPR policy require, or allow, IPR holders to individually disclose their most restrictive licensing terms, including the maximum royalty rates they would charge, prior to the adoption of the standard this will not lead to a restriction of competition within the meaning of Article 101(1) as long as the rules do not allow for the joint negotiation or discussion of licensing terms in particular royalty rates. Such unilateral ex ante disclosures of most restrictive licensing terms would be one way to enable the standard-setting organisation to take an informed decision based on the disadvantages and advantages of different alternative technologies, not only from a technical perspective but also from a pricing perspective.

288. The inclusion in a standard of substitute technologies (i.e., technology which is regarded by users/licensees as interchangeable with or substitutable for another technology, by reason of the technologies' characteristics and intended use and which could, in the present context, be adopted in an alternative standard) may limit inter-technology competition. Where a standard is composed of substitute technologies, the arrangement can in practice amount to foreclosure of competitors by excluding one potentially competing alternative technology from being included in a different standard. As a general rule, the inclusion of substitute technologies in a standard is likely to give rise to restrictive effects on competition within the meaning of Article 101(1).

96 See Case 395/87, Ministère public v Jean-Louis Tournier, [1989] ECR 2521, paragraph 38; Joined

Cases 110/88, 241/88 and 242/88, Francois Lucazeau v SACEM, [1989] ECR 2811, paragraph 33.

EN 73 EN

289. Moreover, standards that are not accessible to third parties (i.e., non-members of the relevant standard-setting organisation) may discriminate or foreclose third parties or segment markets according to their geographic scope of application. Thus, the assessment whether the agreement restricts competition will focus on these elements.

290. As the effectiveness of standardisation agreements is often proportional to the share of the industry involved in setting and/or applying the standard, high market shares held by the parties in the market(s) affected by the standard will not necessarily lead to the conclusion that the standard is likely to give rise to restrictive effects on competition.

Standard terms

291. The establishment and use of standard terms must be assessed in the appropriate economic context and in the light of the situation on the relevant market in order to determine whether the standard terms at issue are likely to give rise to restrictive effects on competition.

292. Whether the establishment and use of standard terms is likely to have restrictive effects on competition within the meaning of Article 101(1) depends, amongst other factors, on the extent to which the parties remain free to use other terms when selling their product to the consumer. In other words, it is important to assess whether the standard terms are non-binding.

293. As long as participation in the actual establishment of standard terms is unrestricted (either by participation in the trade association or directly) and transparent, and the established standard terms are non-binding and effectively accessible for anyone, such agreements do not, in general (subject to the caveats set out in paragraphs 295 and 297-298 below), give rise to restrictive effects on competition.

294. If the use of standard terms is binding, there is a need to assess their impact on product quality, product variety and innovation (in particular if the standard terms are binding on the entire market).

295. Moreover, should the standard terms (binding or non-binding) contain any terms which have a likely negative effect on competition relating to prices, rebates, interest or other parameters influencing the actual sales price (even if they do not directly set the price), they would give rise to restrictive effects on competition within the meaning of Article 101(1).

296. Effectively accessible and non-binding standard terms for the sale of consumer goods or services (on the presumption that they have no effect on price) thus generally do not have any restrictive effect on competition since they are unlikely to lead to any negative effect on product quality, product variety or innovation. There are, however, two general exceptions where a more in-depth assessment would be required.

297. Firstly, standard terms for the sale of consumer goods or services where the standard terms define the scope of the product sold to the customer, and where therefore the risk of limiting product choice is more important, could give rise to restrictive effects on competition within the meaning of Article 101(1). This could be the case when the wide-spread use of the standard terms de facto leads to a limitation of innovation and product variety. For instance, this may arise where standard terms in insurance

EN 74 EN

contracts limit the customer's practical choice of key elements of the contract, such as the standard risks covered. Even if the use of the standard terms is not compulsory, they might undermine the incentives of the competitors to compete on product diversification.

298. Secondly, even if the standard terms do not define the actual scope of the end-product they might be a decisive part of the transaction with the customer for other reasons, thereby raising the same concerns as set out in the previous paragraph. An example would be online shopping where customer confidence is essential (e.g., in the use of safe payment systems, a proper description of the products, clear and transparent pricing rules, flexibility of the return policy, etc). As it is difficult for customers to make a clear assessment of all these elements, they tend to favour widespread practices. Standard terms regarding these elements could therefore rapidly spread within the industry and become a de facto standard as the more customers buy under these standard terms, the more they will be trusted, and then used, by other customers. As a result more and more companies would use these standard terms, and compliance with them could become necessary to sell in the market. Even though non-binding, these standard terms would become a de facto standard, whose effects are very close to a binding standard and need to be analysed accordingly.

299. The higher the common market share of the competitors applying the standard terms, the more likely are restrictive effects on competition in the form of a limitation of product variety.




300. Standardisation agreements can give rise to significant efficiency gains. For example, EU wide standards may facilitate market integration and allow companies to market their goods and services in all Member States, leading to increased consumer choice and decreasing prices. Standards which establish technological interoperability often encourage competition on the merits between technologies from different companies and help prevent lock-in to one particular supplier. Furthermore, standards may reduce transaction costs for sellers and buyers. Standards on for instance quality, safety and environmental aspects of a product may in addition facilitate consumer choice and can lead to increased product quality. Standards also play an important role for innovation. They can reduce the time it takes to bring a new technology to the market and facilitate innovation by allowing companies to build on top of agreed solutions.

301. To materialise those efficiency gains in the case of standardisation agreements, the necessary information to apply the standard must be effectively available to those wishing to enter the market and an appreciable proportion of the industry must be involved in the setting of the standard in a transparent manner. The rules of the standard-setting organisations should contain sufficient safeguards to prevent the standard-setting process from being biased towards one or several participants. The effects on innovation must be analysed on a case by case basis. However, for

EN 75 EN

instance standards creating compatibility on a horizontal level between different technology platforms are considered to be likely to give rise to efficiency gains.

Standard terms

302. The use of standard terms can entail economic benefits such as making it easier for customers to compare the conditions offered and thus facilitate switching between companies. They might also lead to efficiency gains in the form of savings in transaction costs and, in certain sectors (in particular where the contracts are of a complex legal structure), facilitate entry.


303. Restrictions that go beyond what is necessary to achieve the efficiency gains generated by a standardisation agreement or standard terms do not fulfil the criteria of Article 101(3).


304. The assessment of each standardisation agreement must take into account the nature of the standard and its likely effect on the markets concerned, on the one hand, and the scope of restrictions that possibly go beyond the objective of achieving efficiencies, on the other.

305. Standardisation agreements that entrust certain bodies with the exclusive right to test compliance with the standard, or impose restrictions on marking of conformity with standards, unless imposed by regulatory provisions, go beyond the objective of achieving efficiencies and may not be indispensable to the attainment of these objectives.

306. By their nature, standards will not (and should not) include all possible specifications or technologies. In cases where having only one technological solution would benefit consumers or the economy at large this standard must be set on a non-discriminatory basis. Technology neutral standards are presumed to lead to larger efficiency gains. In any event, it must be justifiable why one standard is chosen over another.

307. All competitors in the market(s) affected by the standard should have the possibility to take part in the setting of the standard. Therefore, participation in standard-setting should be open to all of them unless the parties demonstrate significant inefficiencies of such participation or unless recognised procedures are foreseen for the collective representation of interests.

308. As a general rule standardisation agreements should cover no more than what is necessary to ensure their aims, whether this is technical compatibility or a certain level of quality.

309. Standardisation agreements making a standard binding and obligatory for the industry are in principle not indispensable.

Standard terms

EN 76 EN

310. It is generally not justified to make standard terms binding and obligatory for the industry or the members of the trade association that established them. It is, however, not excluded that binding standard terms may in a specific case be indispensable to the attainment of the efficiency gains generated by them.



311. Efficiency gains attained by indispensable restrictions must be passed on to consumers to an extent that outweighs the restrictive effects on competition caused by a standardisation agreement or by standard terms. A relevant part of the analysis of likely pass-on to consumers is which procedures are used to guarantee that the interests of the users of standards are protected. Where standards facilitate interoperability and competition between new and already existing products, services and processes, and secure multiple supply sources it can be presumed that the standard will benefit consumers.

Standard terms

312. Both the risk of restrictive effects on competition and the likelihood of efficiency gains increase with the companies' market shares and the extent to which the standard terms are used. Hence, it is not possible to provide any general "safe harbour" within which there is no risk of restrictive effects on competition or which would allow the presumption that efficiency gains will be passed on to consumers to an extent that outweighs the restrictive effects on competition.

313. However, certain efficiency gains generated by standard terms, such as increased comparability of the offers on the market, are necessarily passed on to the consumers. As regards other possible efficiency gains, such as lower transaction costs, it needs to be assessed on a case-by-case basis and in the relevant economic context whether these are likely to be passed on to consumers.


314. Whether a standardisation agreement affords the parties the possibility of eliminating competition depends on the various sources of competition in the market, the level of competitive constraint that they impose on the parties and the impact of the agreement on this competitive constraint. While market shares are relevant for this analysis, the magnitude of remaining sources of actual competition cannot be assessed exclusively on the basis of market share except in cases where a standard becomes a de facto industry standard. In the latter case competition may be eliminated if third parties are foreclosed from access on fair, reasonable and non-discriminatory terms to this standard. Standard terms used by a majority of the industry might create a de facto industry standard and thus raise the same concerns.

7.5. Examples

315. Setting standards competitors cannot satisfy

Example 1

EN 77 EN

Situation: A standard-setting organisation sets and publishes quality standards that are widely used by the relevant industry. Most members of the industry take part in the setting of the standard. Prior to the adoption of the standard, a new entrant has developed a product that is technically equivalent in terms of the performance and functional requirements, which is recognised by the technical committee of the standard-setting organisation. However, the technical specifications of the quality standard are drawn up in such a way as to not allow for this or other new products to comply with the standard. Analysis: This standardisation agreement is likely to give rise to restrictive effects on competition within the meaning of Article 101(1) and is unlikely to meet the criteria of Article 101(3). Although it is common that quality standards do not include all possible specifications, in this case the members of the standards development organisation have set the standard in such a way that products of their competitors which are based on other technological solutions cannot satisfy it, even though they have equivalent performance. Hence, this agreement, which has not been set on a non-discriminatory basis, will reduce or prevent innovation and product variety. It is unlikely that the way the standard is drafted will lead to greater efficiency gains than a neutral one.

316. A standard-setting organisation without clear rules on IPR

Example 2 Situation: The rules of the standards development organisation X encourage its members to disclose all their IPR (and pending IPR) which are essential to a proposed standard but does not require them do so. The IPR rules of X do not impose an obligation on its members to commit to license the IPR included in a standard. Analysis: Standardisation agreements concluded by X are likely to give rise to restrictive effects on competition within the meaning of Article 101(1) and are unlikely to meet the criteria of Article 101(3). In order not to produce restrictive effects on competition, the parties to a standardisation agreement should ensure that the IPR rules require that companies declare during the standard-setting process (i.e., before the standard is provisionally agreed) that they have (or believe to have) essential IPR (and pending IPR) which may read on a proposed standard, and that these companies identify what essential IPR (and pending IPR) they have (or believe to have) before that standard is formally published. Although there is no general requirement that parties to a standardisation agreement must declare that they are willing to licence their relevant IPR on a fair, reasonable and non-discriminatory basis, failure to do so may restrict competition within the meaning of Article 101(1).

317. Unbinding and transparent standard covering a large part of the market

Example 3 Situation: A number of consumer electronics manufacturers with substantial market shares agree to develop a new standard for a product to follow up the DVD. Analysis: Provided that (a) the manufacturers remain free to produce other new products which do not conform to the new standard, (b) participation in the

EN 78 EN

standard-setting is unrestricted and transparent, and (c) the standardisation agreement does not otherwise restrict competition, then Article 101(1) is not infringed. If the parties agreed to only manufacture products which conform to the new standard, the agreement would limit technical development, reduce innovation and prevent the parties from selling different products, thereby creating restrictive effects on competition within the meaning of Article 101(1).

318. Standards in the insurance sector

Example 4 Situation: A group of insurance companies come together to agree non-binding standards for the installation of certain security devices (i.e., components and equipment designed for loss prevention and reduction and systems formed from such elements) for which there is currently no harmonised EU standard in place. Those non-binding standards may cover one Member State or a number of Member States. The insurers have brought the specific need for such standards to the attention of the relevant EU standards body and are involved with them with a view to putting an EU harmonised standard in place to address this need. The non-binding standards set by the insurance companies (a) are agreed in order to address a specific need and to assist insurers to manage risk and keep insurance premiums low; (b) are discussed with the majority of installers in the affected Member States and their views are taken on board prior to finalisation of the standards; (c) are published by the relevant insurance association(s) on a dedicated section of its (or their) website(s) so that any installer or other interested party can access them easily and (d) will lapse as soon as a harmonised EU standard is adopted. Analysis: The process for setting these standards is transparent and allows for the participation of interested parties. In addition, the result is easily accessible on a reasonable and non-discriminatory basis for anyone that wishes to have access to it. Provided that the standard does not have negative effects on the downstream market (for example by excluding certain installers through very specific and not justified requirements for installations that cannot be fulfilled by certain installers) it is not likely to lead to a restrictive effect on competition. However, even if the standards led to a restrictive effect on competition the conditions set out in Article 101(3) may be fulfilled. The standards would assist insurers to analyse to what extent such installation systems reduce relevant risk and prevent losses so that they can reduce premiums. They would also, subject to the caveat set out above regarding the downstream market, be more efficient for installers, allowing them to comply with one set of standards for all insurance companies in the affected Member States rather than be tested by every insurance company separately. They could also facilitate consumers switching between insurers. In addition, they could be beneficial for smaller insurers who may not have the capacity to test separately. As regards the other conditions of Article 101(3), it seems that the non-binding standards do not go beyond what is necessary to achieve the efficiencies in question; that benefits would be passed on to the consumers (some would even be directly beneficial for the consumers); and that the restrictions would not lead to an elimination of competition. Therefore these standards may fulfil the criteria of Article 101(3) until such time as EU harmonised standards are in place to address the specific need that these standards address.

319. Environmental standards

EN 79 EN

Example 5 Situation: Almost all producers of washing machines agree, with the encouragement of a public body, to no longer manufacture products which do not comply with certain environmental criteria (e.g., energy efficiency). Together, the parties hold 90% of the market. The products which will be thus phased out of the market account for a significant proportion of total sales. They will be replaced with more environmentally friendly, but also more expensive products. Furthermore, the agreement indirectly reduces the output of third parties (e.g., electric utilities, suppliers of components incorporated in the products phased out). Analysis: The agreement grants the parties control of individual production and concerns an appreciable proportion of their sales and total output, whilst also reducing third parties’ output. Product variety, which is partly focused on the environmental characteristics of the product, is reduced and prices will probably rise. Therefore, the agreement is likely to give rise to restrictive effects on competition within the meaning of Article 101(1). The involvement of the public authority is irrelevant for this assessment. However, newer, more environmentally friendly products are more technically advanced, offering qualitative efficiencies in the form of more washing machine programmes which can be used by consumers. Furthermore, there are cost efficiencies for the purchasers of the washing machines resulting from lower running costs in the form of reduced consumption of water, electricity and soap. These cost efficiencies are realised on markets which are different from the relevant market of the agreement. Nevertheless, these efficiencies may be taken into account as the markets on which the restrictive effects on competition and the efficiency gains arise are related and the group of consumers affected by the restriction and the efficiency gains is substantially the same. The efficiency gains outweigh the restrictive effects on competition in the form of increased costs. Other alternatives to the agreement are shown to be less certain and less cost-effective in delivering the same net benefits. Various technical means are economically available to the parties in order to manufacture washing machines which do comply with the environmental characteristics agreed upon and competition will still take place for other product characteristics. Therefore, the criteria of Article 101(3) are fulfilled.

320. Government encouraged standardisation

Example 6 Situation: In response to the findings of research into the recommended levels of fat in certain processed food conducted by a government-funded think tank in one Member State, several major manufacturers of the processed foods in the same Member State agree, through formal discussions at an industry trade association, to set recommended fat levels for the products. Together, the parties represent 70% of sales of the products within the Member State. The parties' initiative will be supported by a national advertising campaign funded by the think tank highlighting the dangers of a high fat content in processed foods. Analysis: Although the fat levels are recommendations and therefore voluntary, as a result of the wide publicity resulting from the national advertising campaign, the recommended fat levels are likely to be implemented by all manufacturers of the processed foods in the Member State. It is therefore likely to become a de facto

EN 80 EN

maximum fat level in the processed foods. Consumer choice across the product markets could therefore be reduced. However, the parties will be able to continue to compete with regard to a number of other characteristics of the products, such as price, product size, quality, taste, other nutritional and salt content, balance of ingredients, and branding. Moreover, competition regarding the fat levels in the product offering may increase where parties seek to offer products with the lowest levels. The agreement is therefore unlikely to give rise to restrictive effects on competition within the meaning of Article 101(1).

321. Non-binding and open standard terms used for contracts with end-users

Example 7 Situation: A trade association for electricity distributors establishes non-binding standard terms for the supply of electricity to end-users ("Standard terms for sale of electricity"). The establishment of the standard terms is made in a transparent and non-discriminatory manner. The standard terms cover issues such as the specification of the point of consumption, the location of the connection point and the connection voltage, provisions on service reliability as well as the procedure for settling the accounts between the parties to the contract (e.g., what happens if the customer does not provide the supplier with the readings of the measurement devices). The standard terms do not cover any issues relating to prices, i.e., they contain no recommended prices or other clauses related to price. Any company active within the sector is free to use the standard terms as it sees fit. About 80% of the contracts concluded with end-users in the relevant market are based on these standard terms. Analysis: These standard terms are not likely to give rise to restrictive effects on competition within the meaning of Article 101(1). Even if they have become industry practice, they do not seem to have any appreciable negative impact on prices, product quality or variety.

322. Standard terms used for contracts between companies

Example 8 Situation: Construction companies in a certain Member State come together to establish non-binding and open standard terms and conditions for use by a contractor when submitting a quotation for construction work to a client. A form of quotation is included together with terms and conditions suitable for building or construction. Together, the documents create the construction contract. Clauses cover such mattes as contract formation, general obligations of the contractor and the client and non-price related payment conditions (e.g., a provision specifying the contractor's right to give notice to suspend the work for non-payment), insurance, duration, handover and defects, limitation of liability, termination, etc. In contrast to example 7, these standard terms would often be used between companies, one active upstream and one active downstream. Analysis: These standard terms are not likely to have restrictive effects on competition within the meaning of Article 101(1). There would normally not be any significant limitation in the customer's choice of the end-product, namely the construction work. Other restrictive effects on competition do not seem likely. Indeed, several of the clauses above (handover and defects, termination, etc.) would

EN 81 EN

often be regulated by law.

323. Standard terms facilitating the comparison of different companies' products

Example 9 Situation: A national association for the insurance sector distributes non-binding standard policy conditions for house insurance contracts. These conditions give no indication of the limit of cover of the risk, the level of insurance premiums or excesses payable by the insured. While the majority of insurance companies use standard policy conditions, not all their contracts contain the same conditions as these are adapted to each client's individual needs and therefore there is no de facto standardisation of insurance products offered to consumers. The standard policy conditions enable consumers and consumer organisations to compare the policies offered by the different insurers. A consumer association is involved in the process of laying down the standard policy conditions. They are also available for use by new entrants, on a non-discriminatory basis. Analysis: These standard policy conditions relate to the composition of the final insurance product. Any limitation in product variety as a result of insurance companies using such standard policy conditions is likely to be outweighed by efficiencies such as facilitation of comparison by consumers of conditions offered by insurance companies. These comparisons in turn facilitate switching between insurance companies and thus enhance competition. The fact that the consumer association has participated in the process could increase the likelihood of efficiencies being passed on to consumers. The standard policy conditions are also likely to reduce transaction costs and facilitate entry for insurers on a different geographic and/or product markets. Moreover, the restrictions do not seem to go beyond what is necessary to achieve the identified efficiencies and competition would not be eliminated. Consequently, the criteria of Article 101(3) are likely to be fulfilled.

en - ec.europa.euec.europa.eu/.../consultations/2010_horizontals/guidelines_en.pdf · en 2 en draft...

Documents